VOLUME 82 · No. 10 · OCTOBER 2016...ECMO management guided by echo: is it feasible? Tritapepe L....

V O L U M E 8 2 · N o. 1 0 · O C T O B E R 2 0 1 6

1039ColloidophobiaRipollés Melchor J., Fries D., Chappell D.

1043ORIGINAL ARTICLESRight ventricle dilation as a prognostic factor in refractory acute respiratory distress syndrome requir-ing veno-venous extracorporeal membrane oxygena-tionLazzeri C., Cianchi G., Bonizzoli M., Batacchi S., Terenzi P., Bernardo P., Valente S., Gensini G. F., Peris A.

1050Comparison of three videolaryngoscopes for double-lumen tubes intubation in simulated easy and difficult airways: a randomized trialEl-Tahan M. R., Al’ghamdi A. A., Khidr A. M., Gaarour I. S.

1059Effects of passive leg raising on microvascular venous compartment in critically ill patientsDe Blasi R. A., Arcioni R., Brancadoro D., Rocco M.

1023EDITORIALSECMO management guided by echo: is it feasible?Tritapepe L.

1026Microvascular fluid cup: sturdy in the healthy, but bottomless in septic shockBenes J., Monnet X.

1029When the nerve block is blocked by local resistanceQu J. Z., Alston T. A.

1032Basics to perform and present statistical analyses in scientific biomedical reports. Part 3Cesana B. M., Cavaliere F.

1036The high nasal flow therapy for pre-oxygenation: a new strategy for conventional procedures in intensive care settings?Terragni P., Cossu A. P.

MINERVA ANESTESIOLOGICA

Vol. 82 October 2016 No. 10

CONTENTS

Vol. 82 - No. 10 MINERVA ANESTESIOLOGICA I

OFFICIAL JOURNAL OF ITALIAN SOCIETY OF ANESTHESIOLOGY, ANALGESIA,RESUSCITATION AND INTENSIVE CARE (SIAARTI)

ITALIAN JOURNAL OF ANESTHESIOLOGY AND ANALGESIAMONTHLY JOURNAL FOUNDED IN 1935 BY A. M. DOGLIOTTI

OFFICIAL JOURNAL OF ITALIAN SOCIETY OF ANESTHESIOLOGY, ANALGESIA,RESUSCITATION AND INTENSIVE CARE (S.I.A.A.R.T.I.)

CONTENTS

II MINERVA ANESTESIOLOGICA October 2016

1069The effects of minimal-dose versus low-dose S-ketamine on opioid consumption, hyperalgesia, and postoperative delirium: a triple-blinded, randomized, active- and placebo-controlled clinical trialBornemann-Cimenti H., Wejbora M., Michaeli K., Edler A., Sandner-Kiesling A.

1077Point-of-care-based protocol with first-line therapy with coagulation factor concentrates is associated with decrease allogenic blood transfusion and costs in cardiovascular surgery: an Italian single-center expe-rienceTrevisan D., Zavatti L., Gabbieri D., Pedulli M., Giordano G., Meli M.

1089Whole-exome sequencing of a family with local anes-thetic resistanceClendenen N., Cannon A. D., Porter S., Robards C. B., Parker A. S., Clendenen S. R.

1098EXPERTS’ OPINIONSHazards of intubation in the ICU: role of nasal high flow oxygen therapy for preoxygenation and apneic oxygenation to prevent desaturationRicard J. D.

1107The risk of infusing gelatin? Die-hard misconceptions and forgotten (or ignored) truthsPisano A., Landoni G., Bellomo R.

1115A dynamic view of dynamic indicesFischer M. O., Guinot P. G., Biais M., Mahjoub Y., Mallat J., Lorne E.

1122LETTERS TO THE EDITORNear-zero difficult tracheal intubation and tracheal intubation failure: too good to be trueCorso R. M., Di Giacinto I., Sorbello M., Piraccini E., Petrini F.

1123Besta Airway Algorithm in morbidly obese patients: expertise and safetyCagnazzi E., Latronico N., Pe F., Mosca A.

1124Delayed CO2 embolism: importance of early postop-erative surveillanceLeoni A., Antonelli A., Pocar M.

1125Cervical emphysema and pneumomediastinum due to isolated pharyngeal perforation after blunt traumaVissani M., Lentischio L., Nicoletta G., Pugliese F., Fedeli C., Zava R.

1127TOP 50 MINERVA ANESTESIOLOGICA REVIEWERS

About the cover: The cover shows a genetic variant associated with local anesthetics resistance in the gene encoding for Nav1.5. Immunohistochemical staining shows strong Nav1.5 immuno-reactivity in peripheral nerves (C), dorsal root ganglia (D), and the brain (E). Cardiac tissue was stained with Nav1.5 antibody in the absence (A) or presence (B) of a block-ing peptide to denote positive and negative controls, respectively. For more information, see article by Clendenen N. et al. on page 1089.

V O L U M E 8 2 · N o. 1 0 · O C T O B E R 2 0 1 6

EDITORIAL BOARD

EDITOR IN CHIEFF. CavaliereRoma, Italy

General Critical CareAssociate Editor

G. M. Albaiceta (Oviedo, Spain)Section Editor

G. Biancofiore (Pisa, Italy)E. De Robertis (Napoli, Italy)

Circulation Critical CareSection Editor

S. Scolletta (Siena, Italy)E. Bignami (Milano, Italy),

Respiration Critical CareSection Editor

S. Grasso (Bari, Italy)P. Terragni (Sassari, Italy),

Neurocritical CareSection Editor

F. S. Taccone (Brussels, Belgium)

Section EditorA. Giannini (Milano, Italy)

Section EditorM. Allegri (Parma, Italy)F. Coluzzi (Roma, Italy)

ETHICS PAIN

Section EditorB. M. Cesana (Brescia, Italy)

MEDICAL STATISTIC

General AnesthesiaAssociate Editor

M. Rossi (Roma, Italy)Section Editor

E. Cohen (New York, USA)P. Di Marco (Roma, Italy)

J. T. Knape (Utrecht, The Netherlands)O. Langeron (Paris, France)

P. M. Spieth (Dresden, Germany)

Pediatric AnesthesiaSection Editor

M. Piastra (Roma, Italy)

Obstetric AnesthesiaSection Editor

E. Calderini (Milano, Italy)

Regional AnesthesiaSection Editor

A. Apan (Giresun, Turkey)M. Carassiti (Roma, Italy)

CRITICAL CARE ANESTHESIA

MANAGING EDITORA. Oliaro

Torino, Italy

MINERVA ANESTESIOLOGICAITALIAN JOURNAL OF ANESTHESIOLOGY AND ANALGESIA

MONTHLY JOURNAL FOUNDED IN 1935 BY A. M. DOGLIOTTIOFFICIAL JOURNAL OF ITALIAN SOCIETY OF ANESTHESIOLOGY, ANALGESIA,

RESUSCITATION AND INTENSIVE CARE (S.I.A.A.R.T.I.)

Vol. 82 - No. 10 MINERVA ANESTESIOLOGICA III

This journal is PEER REVIEWED and is quoted in:

Current Contents, SciSearch, PubMed/MEDLINE, EMBASE, Scopus

G E N E R A L I N F O R M A T I O N

Subscribers: Payment to be made in Italy: a) by check; b) by bank transfer to: Edizioni Minerva Medica, INTESA SANPAOLO Branch no. 18 Torino. IBAN: IT45 K030 6909 2191 0000 0002 917 c) through postal account no. 00279109 in the name of Edizioni Minerva Medica, Corso Bramante 83-85, 10126 Torino; d) by credit card Diners Club International, Master Card, VISA, American Express. Foreign countries: a) by check; b) by bank transfer to: Edizioni Minerva Medica, INTESA SANPAOLO Branch no. 18 Torino. IBAN: IT45 K030 6909 2191 0000 0002 917; BIC: BCITITMM c) by credit card Diners Club International, Master Card, VISA, American Express. Members: for payment please contact the Administrative Secretariat SIAARTI.Notification of changes to mailing addresses, e-mail addresses or any other subscription information must be received in good time. Notification can be made by sending the new and old information by mail, fax or e-mail or directly through the website www.minerva-medica.it at the section “Your subscriptions - Contact subscriptions department”.Complaints regarding missing issues must be made within six months of the issue’s publication date.Prices for back issues and years are available upon request.© Edizioni Minerva Medica - Torino 2016. All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior permission of the copyright owner.Monthly publication. Authorized by Turin Court no. 281 of 24-1-1953.

Published byEdizioni Minerva Medica - Corso Bramante 83-85 - I-10126 Turin (Italy)

Tel. +39 (011) 678282 - Fax +39 (011) 674502 Website: www.minervamedica.it

Editorial office: [email protected] - Subscriptions: [email protected]: [email protected]

Presidential Office: Antonio Corcione, UOC Anestesia e TIPO, Azienda Ospedaliera dei Colli, V. Monaldi, Via Leonardo Bianchi 1, 80131 Naples, Italy - Tel. 081 7065214. E-mail: [email protected]

Technical Secretariat: S.I.A.A.R.T.I., Viale dell’Università 11, 00185 Rome, Italy - Tel. 06 4452816

E-mail: [email protected] - Website: www.siaarti.it

Administrative Secretariat S.I.A.A.R.T.I. - Edizioni Minerva Medica - Corso Bramante 83-85 - 10126 Turin, Italy - Tel. +39 (011) 678282 - Fax +39 (011) 674502 - E-mail: [email protected]

ITALY

Individual: Print € 130,00 Print+Online € 135,00 Online € 140,00

Institutional: Print € 180,00

Online Small € 350,00 Medium € 407,00 Large € 429,00

Print+Online Small € 367,00 Medium € 420,00 Large € 446,00

EUROPEAN UNION





OUTSIDE THEEUROPEAN UNION





ANNUAL SUBSCRIPTIONS

IV MINERVA ANESTESIOLOGICA October 2016

Impact factor: 2.036

This Journal complies with the Code of Self-Discipline of Medical/Scientific Publishers associated with FARMAMEDIA and may accept advertisingThis Journal is associated with

Vol. 82 - No. 10 MiNerVa aNestesiologica 1023

an increased afterload of the right chambers.3 The first step in the treatment of ACP in ARDS patients is the reduction of pulmonary artery hypertension, to reduce the incidence of the right ventricular dysfunction. This has been dramatically reduced by the implementation of protective lung ventilation, but still remains as high as 25%.4, 5 In case of failure of this treatment, we have to choose other measure of support to protect both the lung and the right ventricle, i.e. the VV-ecMo.

The institution of VV-ECMO leads to the resolution of hypoxemia and hypercapnia, and to a reduction of airway pressures, which results in decreased pulmonary vascular re-sistance. This may reverse the hemodynamic instability associated with the right ventricular dysfunction, as described by Lazzeri C et al.1

It is very important to assess the right ven-tricular function day by day with the echocar-diography evaluation.

Because of the high acoustic impedance caused by high PEEP values, the transthoracic echocardiography (TTE) may not be feasible. Transesophageal echocardiography (TEE) is preferred because it is easy to perform in the ARDS patients, who are often intubated and sedated.

A recently published position paper on the management of the ecMo recommends that an echocardiography-trained physician should be part of the team caring for patients on ECMO.6

in this issue of Minerva Anestesiologica, the manuscript by Lazzeri C et al.1 focuses on

echocardiographic assessment of the right ven-tricle in ECMO patients suffering from acute respiratory distress syndrome (ARDS), trying to highlight the predictors of mortality.

Why does echocardiography play an impor-tant role in risk stratification of patients under-going extracorporeal membrane oxygenation for severe respiratory failure?

In case of ARDS, after an initial aggressive treatment, the patient can have a worsening of their respiratory insufficiency.

Simultaneously he presents high values of central venous pressure and dilation of inferior vena cava (IVC) with fluid retention both in pleural and in abdomen cavities.2 Why does that happen?

If we perform an echocardiography evalua-tion, we can show a dilation of the right ven-tricle with associated pulmonary hypertension. This picture is well described in literature and is called Acute Cor Pulmonale (ACP). When this situation is present, the therapeutic strat-egy changes rapidly and it is mandatory to achieve a compromise between the best venti-latory positive end-expiratory pressure (PEEP) and circulatory stability. The therapeutic goal is to use protective ventilation finalized at re-ducing the right ventricular failure related to

E D I T O R I A L

ECMO management guided by echo: is it feasible?Luigi TRITAPEPE

UOD Anesthesia and Intensive Care in Cardiac Surgery, Department of Cardiovascular, Respiratory, Nephrologic, Anesthesiologic and Geriatric Sciences, Sapienza University of Rome, Rome, ItalyCorresponding author: Luigi Tritapepe, Department of Cardiovascular, Respiratory, Nephrologic, Anesthesiologic and Geriatric Sci-ences, Sapienza University of Rome, Rome, Italy. E-mail: [email protected]

Anno: 2016Mese: OctoberVolume: 82No: 10Rivista: Minerva anestesiologicaCod Rivista: Minerva Anestesiol

Lavoro: 11377-MAStitolo breve: ECMO MANAGEMENT GUIDED BY ECHOprimo autore: TRITAPEPEpagine: 1023-5citazione: Minerva Anestesiol 2016;82:1023-5

Comment on p. 1043.

Minerva Anestesiologica 2016 October;82(10):1023-5© 2016 EDIZIONI MINERVA MEDICAOnline version at http://www.minervamedica.it

TRITAPEPE ECMO MANAGEMENT GUIDED BY ECHO

1024 MiNerVa aNestesiologica october 2016

the femoral vein, advanced into the IVC, and in the internal jugular vein, advanced into the superior vena cava (SVC). Alternatively, we have to cannulate both femoral veins choos-ing to re-inject blood in the cannula positioned in the right atrium. Another possible solution is the use of a double-lumen cannula inserted through the right internal jugular vein until the SVC. In this stage, TEE is of paramount importance to avoid any complication during cannula insertion. An Echo Guided insertion of the cannulas is mandatory and visualization of guidewire through a mid-esophageal bica-val view allows the contemporary view of the IVC, SVC, tricuspid valve, right atrium and atrial septum.12 It is very important to check the guidewire in both Cavae to avoid any com-plicated position, i.e. through tricuspid valve into the right ventricle or into the coronary sinus or through a PFO or even worse in the pericardial space.

The echocardiographic control of the tip of the venous cannula is of paramount impor-tance to avoid keeping it too close to the atrial wall or to the other venous cannula and to re-duce drainage and produce recirculation with the increase of shunt and subsequent hypoxia. With the echo TEE control, we can achieve an optimal drainage and avoid recirculation.13 Finally the echo view provides important in-formation to prevent the exclusion of hepatic venous return due to a malposition of inferior cannula. The study of Doppler hepatic vein flow is highly informative in this sense.14, 15 in addition, the echo control allows us to change the position of the cannula and evaluate how the degree of the patient’s blood volume could affect the performance of the ECMO.

Echocardiography plays a crucial role dur-ing every step of ECMO support. It helps the correct positioning of the cannula, and allows a continuous control of the efficacy of the drain-age and of the oxygenation. It is evident that echocardiography support cannot lead to an increase of survival in ECMO patient but, as shown by Lazzeri C et al.,1 allows us to stratify the risk of patients with ARDS and to avoid major complications.16

However, the role of echocardiography in ECMO is not widely accepted and is still poor-ly described in the literature.

Echocardiographic examination of the right ventricle requires a long-axis and a short-axis views to evaluate the size of the cavity, the left ventricle relationship and septal kinetics. The assessment can be completed by the Doppler examination of the right ventricular ejection flow and of tricuspid regurgitation when it is present, to measure the systolic pulmonary artery pressure. Measure of TAPSE (tricuspid annular plane systolic excursion) is simple and useful from the point of view of prognosis, and avoids the need to measure the right ventricu-lar FAC (fractional area change), more difficult to assess. In addition, TDI (Tissue Doppler Im-aging) of the right ventricle is very useful for measuring the diastolic and systolic function in only few minutes.7 In literature, authors de-fined moderate right ventricular dilatation as a ratio greater than 0.6 and major right ventricu-lar dilatation as a ratio greater than or equal to 1.8 When the right ventricular dilatation is associated with paradoxical septal motion at end-systole, it reflects the systolic overload of the right ventricle, and the systolic eccentricity index has been proposed to “quantify” the sys-tolic overload of the right ventricle.9

Pulmonary hypertension is usually associ-ated with tricuspid regurgitation, but it also depends on right ventricular systolic function, and its value can be surprisingly low when as-sociated with low cardiac output.10 another sign of right ventricular remodeling in ARDS is the thickness of free wall, related to an in-crease in afterload in mechanically ventilated patients.6 Most important is also the detection of a patent foramen ovale (PFO) that can com-plicate the oxygenation of ARDS patients.11 The displacement of septum due to the right ventricular dilatation causes the left ventricu-lar hypo-diastolic status, with a consequent low cardiac output syndrome related to the difficult preload of the left ventricle. This is considered an echo-evaluation of right ven-tricular function in pre-ECMO stage. When the physicians think to institute a VV-ECMO, they need to insert two cannulas, typically in

ECMO MANAGEMENT GUIDED BY ECHO TRITAPEPE


ventricular unloading after initiation of VV ECMO. Am J Respir Crit Care Med 2015;191:346-7.

9. Arkles JS, Opotowsky AR, Ojeda J, Rogers F, Liu T, Prassana V, et al. Shape of the right ventricular Doppler envelope predicts hemodynamics and right heart function in pulmonary hypertension. Am J Respir Crit Care Med 2011;183:268-76.

10. Bouferrache K, Vieillard-Baron A. Acute respiratory distress syndrome, mechanical ventilation, and right ven-tricular function. Curr Opin Crit Care 2011;17:30-5.

11. Conrad SA, Grier LR, Scott LK, Green R, Jordan M. Per-cutaneous cannulation for extracorporeal membrane oxy-genation by intensivists: a retrospective single-institution case series. Crit Care Med 2015;43:1010-5.

12. Hahn RT, Abraham T, Adams MS, Bruce CJ, Glas KE, Lang RM, et al. Guidelines for performing a compre-hensive transesophageal echocardiographic examination: recommendations from the American Society of Echo-cardiography and the Society of Cardiovascular Anesthe-siologists. J Am Soc Echocardiogr 2013;26:921-64.

13. Abrams D, Bacchetta M, Brodie D. Recirculation in venovenous extracorporeal membrane oxygenation. ASAIO J 2015;61:115-21.

14. Dolch ME, Frey L, Buerkle MA, Weig T, Wassilowsky D, Irlbeck M. Transesophageal echocardiography-guided technique for extracorporeal membrane oxygenation du-al-lumen catheter placement. ASAIO J 2011;57:341-3.

15. Chimot L, Marque S, Gros A, Gacouin A, Lavoué S, Ca-mus C, et al. Avalon(c) bicaval dual-lumen cannula for venovenous extracorporeal membrane oxygenation: sur-vey of cannula use in France. ASAIO J 2013;59:157-61.

16. Taylor MA, Taylor BS. Cardiac ultrasound and extracor-poreal life support: the two go together. J Am Soc Echo-cardiogr 2015;28:A18-9.

References 1. Lazzeri C, Cianchi G, Bonizzoli M, Batacchi S, Terenzi

P, Bernardo P, et al. Right ventricle dilation as a prog-Right ventricle dilation as a prog-nostic factor in refractory ARDS requiring VV-ECMO. Minerva Anestesiol 2016;82:1043-9.

2. Vieillard-Baron a, Matthay M, Teboul JL, Bein T, Schultz M, Magder S, et al. Experts’ opinion on man-agement of hemodynamics in ARDS patients: focus on the effects of mechanical ventilation. intensive care Med 2016;42:739-49.

3. Repessé X, Charron C, Vieillard-Baron A. Right ven-tricular failure in acute lung injury and acute respiratory distress syndrome. Minerva Anestesiol 2012;78:941-8.

4. Boissier F, Katsahian S, Razazi K, Thille AW, Roche-Campo F, Leon R, et al. Prevalence and prognosis of cor pulmonale during protective ventilation for acute respiratory distress syndrome. Intensive Care Med 2013;39:1725-33.

5. Mekontso Dessap A, Boissier F, Charron C, Bégot E, Repessé X, Legras A, et al. Acute cor pulmonale during protective ventilation for acute respiratory distress syn-drome: prevalence, predictors, and clinical impact. Inten-sive Care Med 2016;42:862-70.

6. Combes A, Brodie D, Bartlett R, Brochard L, Brower R, Conrad S, et al. Position paper for the organization of ex-tracorporeal membrane oxygenation programs for acute respiratory failure in adult patients. Am J Respir Crit Care Med 2014;190:488-96.

7. Tritapepe L, De Santis V, Pacilli M. The Right Ventricle and Pulmonary Artery. In: Sarti A, Lorini FL, editors. Echocardiography for Intensivists. Milan: Springer-Ver-lag; 2012. p. 91-97.

8. Miranda DR, van Thiel R, Brodie D, Bakker J. Right

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: June 21, 2016. - Manuscript accepted: June 17, 2016. - Manuscript revised: June 14, 2016. - Manuscript received: April 10, 2016.(Cite this article as: Tritapepe L. ECMO management guided by echo: is it feasible? Minerva Anestesiol 2016;82:1023-5)

1026 Minerva anestesiologica october 2016

study in septic patients the PLR guided fluid management was associated with a significant trend towards better outcomes.7 But still there is much to learn about the fluid management in critically ill patients. one of the unanswered questions is the behavior of microvascular bed after volume loading and its potential limita-tion.

in this issue of Minerva Anestesiologica, an experimental study by De Blasi et al. has been published.8 Using an innovative meth-odology combining the graduated venular occlusion and near infrared spectroscopy the authors managed to obtain important data re-garding the behavior of the microvascular bed in the forearm of critically ill patients and con-trol subjects under Plr testing. this complex (but well described) methodology enabled the authors to estimate not only the stressed and unstressed volumes, but also pressure within the vasculature and microvessel wall tone and flow. A direct comparison between macrohemodynamic indices and microvascu-lar parameters at the bedside was performed. expectedly, the authors found differences be-tween the microvascular bed parameters of healthy controls and those with critical illness. in healthy controls, the Plr-induced volume shift led to positive response in macrohemody-namic indices (cardiac output, stroke volume)

Fluid management in critically ill patients may be extremely difficult. It has been

repeatedly demonstrated, that both under-re-suscitated and over-resuscitated patients have worse outcomes than those with adequate vol-ume management.1 Especially in first phases of critical illness it may be of extreme importance to individually tailor the intravenous volume management. it has been proposed by many experts, that dynamic indices should be used to predict cardiovascular system’s reaction to fluid.2 these methods of preload reserve test-ing, or fluid responsiveness, have gained large popularity among treating physicians and even the most recent ssc guidelines which has been modulated in light of recent evidence recommend the repeated fluid responsiveness assessment.3 Passive leg raising (Plr) is a very practical and easy to perform tool to test the preload reserve. When performed in a ra-tional way as described in a recent editorial,4 it enables the assessment of fluid responsive-ness of critically ill patients with high sensi-tivity and specificity.5, 6 Unlike other methods for the prediction of fluid responsiveness, the Plr test is free of many limitations including spontaneous breathing activity or cardiac ar-rhythmias. in a recently published intervention

E D I T O R I A L

Microvascular fluid cup: sturdy in the healthy, but bottomless in septic shock

Jan Benes 1 *, Xavier Monnet 2

1Department of Anesthesia and Intensive Care Medicine, University Hospital and Faculty of Medicine in Plzeň, Charles University Prague, Plzeň, Czech Republic; 2intensive care Unit, inserm UMr_s 999, Paris-sud University Hospitals, le Kremlin-Bicêtre, France*Corresponding author: Jan Benes, Department of Anesthesia and Intensive Care Medicine, University Hospital and Faculty of Medi-cine in Plzeň, Charles University Prague, Alej Svobody 80, 323 00, Plzeň, Czech Republic. E-mail: [email protected]


Lavoro: 11407-MAStitolo breve: MICROVASCULAR FLUID CUPprimo autore: BENESpagine: 1026-8citazione: Minerva Anestesiol 2016;82:1026-8

comment on p. 1059.

Minerva Anestesiologica 2016 October;82(10):1026-8© 2016 eDiZioni Minerva MeDicaOnline version at http://www.minervamedica.it

MicrovascUlar FlUiD cUP Benes

vol. 82 - no. 10 Minerva anestesiologica 1027

clinical perspective, this study brings some information that supports our contemporary view on the disease of microcirculation in critically ill patients. it has been proved by many and nicely reviewed by the panel of aDQi Workgroup 9 that vascular content is merely one of the players in the playground of acute hemodynamic instability. especially in patients with systemic inflammation, de-rangements in vascular and even more im-portantly venular tone may play an important role. Besides, vascular barrier breakdown with increased leakage of the circulating fluid volume into the interstitium generously con-tributes to the final picture. Hence, increas-ing circulating volume in order to promote venous return and cardiac output may depend not only on the mechanical activity of the pump and Frank-starling forces, but also on the ability of the vasculature to deal with the fluids infused. According to data presented by De Blasi et al.,8 the Plr-induced increase in unstressed volume and pressures within the vasculature of septic shock patients nicely demonstrate the risks of increasing fluid load-ing. in some instances, the Plr was neither associated with increased venous flow (and hence venous return) nor with improvement of cardiac output, but was associated with increased postcapillary pressures potentially leading to higher extravasation and edema formation. However, it still remains debat-able whether other interventions may have better effect. norepinephrine, a possible and reasonable alternative to overcome in part the hemodynamic instability by either increasing the vasomotor tone and preload,10 was admin-istered in patients with septic shock in this study. Decreased microvascular compliance and increase in intravascular pressures were observed among these patients 8 supporting previous clinical observations,11 but because norepinephrine was not part of the study in-tervention these observations are only hypo-thetically linked. in addition, this effect on compliance and pressures is associated with risk of promoting hydrostatic forces driven extravasation. other interventions (other cat-echolamines, inodilators) may hypothetically

in a majority of participants. on the contrary, no important changes were observed in the be-havior of the microvascular bed, the stressed and unstressed volumes as well as pressures in the vasculature and its compliance remained stable after Plr maneuver.

in the critically ill patients, different be-havior of macro- and microvasculature was described, but the most important differences were in those having septic shock. in these patients, volume shifts induced a significant increase of the microvascular bed volume, mostly of the unstressed volume, accompanied by an increase of pressures within and a loss of compliance. in combination with the absence of changes in macrohemodynamic parameters, this suggests derangement of vasomotor tone. in patients with sepsis without shock or with other critical illness, the changes were not that marked and volume changes were directed mostly to the stressed volume compartment. a higher proportion of volume responders and lower incidence of norepinephrine support was observed among these patients.

there are several limitations to this study. First of all, the very small number of subjects and their heterogeneity preclude the authors to draw any strong conclusions. Unlike the con-trol group which has been almost entirely fluid responsive, the number of critically ill patients with preload reserve was rather low, making the extrapolations even more difficult. Besides the cardiovascular system of about a half of the critically ill patients (and all those in septic shock) was supported by norepinephrine, what may mitigate the natural behavior of diseased microvascular bed. the methodology of the study is complex and difficult to reapply by others. some of the results obtained were quite difficult to explain: For instance, the fact that in healthy subjects Plr increased cardiac out-put without concomitantly changing stressed volume, pressures and even flow within the vasculature. last but not least, the forearm skeletal muscle is on the one hand accessible, but may behave in a quite different way from other (and more important) microvascular beds within the body.

Beyond its numerous limitations, from a

Benes MicrovascUlar FlUiD cUP


SiteCollectionDocuments/SSC_Bundle.pdf [cited 2016, aug 31].

4. Monnet X, Teboul JL. Passive leg raising: five rules, not a drop of fluid! Crit Care 2015;19:18.

5. cherpanath tgv, Hirsch a, geerts BF, lagrand WK, Leeflang MM, Schultz MJ et al. Predicting Fluid respon-siveness by Passive Leg Raising: A Systematic Review and Meta-analysis of 23 clinical trials. crit care Med 2016;44:981-91.

6. Monnet X, Marik P, teboul Jl. Passive leg raising for predicting fluid responsiveness: a systematic review and meta-analysis. Intensive Care Med 2016 Jan 29. [Epub ahead of print]

7. richard Jc, Bayle F, Bourdin g, leray v, Debord s, Delannoy B et al. Preload dependence indices to titrate volume expansion during septic shock: a randomized controlled trial. Crit Care 2015;19:5.

8. De Blasi. effects of passive leg raising on microvascu-lar venous compartment in critically ill patients. Minerva Anestesiol 2016;82:1059-68.

9. chawla ls, ince c, chappell D, gan tJ, Kellum Ja, Mythen M et al. vascular content, tone, integrity, and haemodynamics for guiding fluid therapy: a conceptual approach. Br J Anaesth 2014;113:748-55.

10. Monnet X, Jabot J, Maizel J, richard c, teboul Jl. nore-pinephrine increases cardiac preload and reduces preload dependency assessed by passive leg raising in septic shock patients. Crit Care Med 2011;39:689-94.

11. Persichini r, silva s, teboul Jl, Jozwiak M, chemla D, richard c et al. effects of norepinephrine on mean sys-temic pressure and venous return in human septic shock. Crit Care Med 2012;40:3146-53.

help to improve the microvasculature in criti-cally ill patients; however, they are not able to reverse shock state when administered alone. In order to do this, fluid infusion and catechol-amines are still the mainstay even though it may be difficult to find the balance between them. in the future, studies like the one by De Blasi et al. may help us gain a better under-standing of the effect of fluids and vasoactive substances on the microvascular bed.

References

1. Murphy cv, schramm ge, Doherty Ja, reichley rM, gajic o, afessa B et al. The importance of fluid man-agement in acute lung injury secondary to septic shock. Chest. 2009;136:102-9.

2. cecconi M, De Backer D, antonelli M, Beale r, Bakker J, Hofer c et al. consensus on circulatory shock and hemodynamic monitoring. task force of the european society of intensive care Medicine. intensive care Med 2014;40:1795-815.

3. ssc executive committee. sepsis surviving campaign - Updated Bundles in Response to New Evidence; 2015 [Internet]. Available from: www.survivingsepsis.org/

Funding.—The scientific work of Jan Benes is supported by the Charles University Research Fund (project no. P36).Conflicts of interest.—Jan Benes is a past advisory Board member of edwards lifesciences inc., and still in ongoing long-term co-operation with edwards lifesciences inc. and cnsystems. Xavier Monnet is a member of the Medical advisory Board of PUlsion Medical systems (MaQUet-group).Article first published online: May 17, 2016. - Manuscript accepted: May 16, 2016. - Manuscript received: April 27, 2016.(Cite this article as: Benes J, Monnet X. Microvascular fluid cup: sturdy in the healthy, but bottomless in septic shock. Minerva Anestesiol 2016;82:1026-8)


form of gene SCNA5 is carried and expressed in roughly 0.5% of people.4 of course, most carriers have one copy of the ordinary allele and one copy of the a572D allele.

expression of SCN5A as NaV1.5 is promi-nent in the heart, and variations of that gene can predispose to cardiac arrhythmia.5 How-ever, it is not clear that the a572D variant is arrhythmogenic.4 in our institution, local an-esthesia resistance is vanishingly rare among patients presenting for electrophysiology pro-cedures. in our informal survey, no resistance cases were identified.

genetic variation of Vgscs has been linked to congenital pain disorders.6, 7 isoforms 1.7, 1.8, and 1.9 are the prominent ones expressed in the peripheral nervous system.8 clendenen et al. find that SCNA5 is also expressed in pe-ripheral nerves, and they further find that its a572D variant allele coincides with resistance to local anesthesia in three family members.3 it is interesting that the patients did not have cardiac conduction issues (such as long Qt syndrome) even though the affected sodium channel (NaV1.5) is expressed more robustly in the heart than in peripheral nerves.4

since clendenen et al. find that NaV1.5 is expressed in peripheral nerves, that channel plausibly participates in pain initiation and propagation. elsewhere, a patient is reported to concomitantly suffer both a complex re-gional pain syndrome and resistance to local anesthesia.9

local anesthetics primarily work by inhibit-ing voltage-gated sodium channels (Vg-

scs). the drug molecules are often applied to nerves in pharmacologically massive concen-trations. For instance, 1.5% mepivacaine cor-responds to 61 mmol/l. even so, resistance to local anesthetics is sometimes seen,1, 2 and clendenen et al. examined a dramatic case in this issue of Minerva Anestesiologica.3 the pa-tient and two similarly resistant relatives proved to have a point mutation in a specific protein.

there are nine Vgscs present on the surfaces of excitable cells. the channels are numbered NaV1.1 through NaV1.9. there are regulatory beta subunits, but each channel is comprised mostly by a long protein called the alpha sub-unit. this loops back and forth across the cell membrane and forms a voltage-sensitive ion pore. such pores enable axons to conduct elec-trochemical waves along their lengths.

in the three patients, the amino acid ala-nine is replaced by aspartate in position 572 of the alpha subunit of the Vgsc known as NaV1.5. the protein is a chain of 2016 amino acids. the gene encoding that protein is called SCN5A and is located on chromosome 3. ala-nine and aspartate are known as a and D, so the substitution is called a572D. Mutation is probably not the most apt word for the widely circulating variant or polymorphism. there are demographic differences, but the a572D

E D I T O R I A L

When the nerve block is blocked by local resistanceJason Z. QU, theodore a. alstoN *

Department of anesthesia, critical care and Pain Medicine, Massachusetts general Hospital, Harvard Medical school, Boston, Ma, Usa*corresponding author: theodore a. alston, Department of anesthesia, critical care and Pain Medicine, Massachusetts general Hospital, 55 Fruit street, Boston, Ma 02114-2696, Usa. e-mail: [email protected]

anno: 2016Mese: octoberVolume: 82No: 10rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

lavoro: 11484-Mastitolo breve: NerVe BlocK BlocKeD BY local resistaNceprimo autore: QUpagine: 1029-31citazione: Minerva anestesiol 2016;82:1029-31

comment on p.1089.

Minerva anestesiologica 2016 october;82(10):1029-31© 2016 eDiZioNi MiNerVa MeDicaonline version at http://www.minervamedica.it

QU NerVe BlocK BlocKeD BY local resistaNce


egies to overcome the problem. For instance, some anesthetics may be less prone than oth-ers to be thwarted by a given genetic variation. Perhaps some variant genes call for relatively high concentrations of an anesthetic.12 genetic resistance is potentially pH-dependent, and alkalization with NaHco3 may be helpful in some cases. Theoretically, a modified local an-esthesia molecule might overcome resistance, but the authors point out that traditional local anesthetics are promiscuous in binding to so-dium channels.3

locals interact with a spectrum of recep-tors, including K+ and ca2+ channels as well as hyperpolarization-activated cyclic nucleotide-gated (HcN) channels.11, 13-15 it is noteworthy that clendenen et al. also discovered a vari-ant form of the KCNE2 gene encoding a K+ channel in the three related patients. Further research may implicate channels other than Vgscs in local anesthetic resistance.

in the laboratory, an inside-the-pore recep-tor site for local anesthetics was inferred in part through the artificial creation of mutant channels, including drug resistant ones.11 It is interesting that some artificial insecti-cides function as sodium channel blockers, and the targeted insects have been changing their sodium channels accordingly.16 sodium channel blockers also occur in nature. an example is tetrodotoxin. as expected, some predators have evolved altered channels in response to tetrodotoxin-wielding prey.17 it is intriguing to wonder if there are any selec-tive pressures favoring variant sodium chan-nels in humans.

For selective pain therapy, inhibitors of NaV1.7 are the main focus of present research. they include small molecules and also pep-tides based on a tarantula venom protein.18, 19 However, the Clenenden findings suggest that it is important to block NaV1.5 in acute pain.3

References

1. Black Wr, Bagot Walters ga. “rachi-resistance” and spinal anaesthesia. Br Med J 1937;1:218-9.

2. Kamsler PM, ruben Je. Procaine resistance; report of a case. anesthesiology 1948;9:313-5.

The Clendenen finding raises many ques-tions. Perhaps the most important one is wheth-er the a572D variant of NaV1.5 is sufficient to cause clinical resistance to local anesthesia. if not, is another factor also required? Overall, what percentage of cases of resistance is caused by the a572D variation or by other variations in NaV1.5? Of note, variation F1737A in NaV1.7 is associated with local anesthesia resistance.10 How often is resistance caused by variations in other VGSCs?10 are there genetic mechanisms of resistance that do not involve VGSCs? What happens to homozygous patients who have two copies of the a572D allele and no copies of the common allele?

it will be important to show whether or not the a572D substitution actually confers drug resis-tance on NaV1.5. this is a delicate assay that is perhaps best done with the isolated protein.11 the authors plausibly propose that “the A572D NaV1.5 mutation results in increased probabil-ity of membrane depolarization in peripheral nerves despite exposure to local anesthetics”.3

it is easy to imagine how a variation of a predominant Vgsc could engender local an-esthetic resistance. it is puzzling that a varia-tion of NaV1.5 would completely block local anesthetic function if that isoform comprises only a minor component of the Vgsc array of peripheral nerves. Perhaps the a572D varia-tion of NaV1.5 perturbs the regulation of the relative expressions of the various types of so-dium channels in different tissues. the compo-sition of Vgscs in a particular subset of the fibers in a nerve sheath may determine clinical drug resistance.11 For instance, so-called C-fi-bers especially contribute to dull-quality pain, and these are 2-4 times less sensitive than a-fibers to anesthetics.11

it is challenging to evaluate a patient’s his-tory of apparent local anesthesia resistance. Skin testing for anesthetic efficacy is possible, but there is no standardized clinical method for doing so. The Clendenen finding points to the possibility of clinically useful genomic assays for both preemptive and post facto diagnosis of this clinical problem.

Identification of genetic mechanisms of local anesthetic resistance may help to devise strat-

NerVe BlocK BlocKeD BY local resistaNce QU


13. Wolf M, schnobel-ehehalt r, Muhling J, Wigand Ma, olschewski. Mechanisms of lidocaine’s action on sub-types of spinal dorsal horn neuron subject to the diverse roles of Na+ and K+ channels in action potential genera-tion. anesth analg 2014;119:463-70.

14. lee Ys, Maruyama M, chang Pc, Park HW, rhee Ks, Hsieh Yc, et al. ryanodine receptor inhibition potenti-ates the activity of Na channel blockers against sponta-neous calcium elevations and delayed after depolariza-tion in langendorff-perfused rabbit ventricles. Heart rhythms 2012;9:1125-32.

15. Zhou c, Ke B, Zhao Y, liang P, liao D, li t, et al. Hy-perpolarization-activated cyclic nucleotide-gated chan-nels may contribute to regional anesthetic effects of lido-caine. anesthesiology 2015;122:606-18.

16. Jiang D, Du Y, Nomura Y, Wang X, Wu Y, Zhorov Bs, et al. Mutations in the transmembrane helix s6 of domain iV confer cockroach sodium channel resistance to so-dium channel blocker insecticides and local anesthetics. insect Biochem Mol Biol 2015;66:88-95.

17. Feldman cr, Brodie eD Jr, Brodie eD 3rd, Pfrender Me. genetic architecture of a feeding adaptation: garter snake (Thamnophis) resistance to tetrodotoxin bearing prey. Proc Biol sci 2010;277:3317-25.

18. Focken t, liu s, chahal N, Dauphinais M, grimwood Me, chowdhury s, et al. Discovery of aryl sulfonamides as isoform-selective inhibitors of NaV1.7 with efficacy in rodent pain models. acs Med chem lett 2016;7:277-82.

19. shcherbatko a, rossi a, Foletti D, Zhu g, Bogin o, galindo casas M, et al. engineering highly potent and selective microproteins against NaV1.7 sodium channel for treatment of pain. J Biol chem 2016;291:13974-86.

3. clendenen N, cannon aD, Porter s, robards cB, Parker AS, Clendenen SR. Whole-exome sequencing of a fam-ily with local anesthetic resistance. Minerva anestesiol 2016;82:1089-97.

4. tester DJ, Valdivia c, Harris-Kerr c, alders M, salis-bury Ba, Wilde aa, et al. epidemiologic, molecular, and functional evidence suggest a572D-scN5a should not be considered an independent lQt3-susceptibility muta-tion. Heart rhythm 2010;7:912-9.

5. lei M, Huang cl, Zhang Y. genetic Na+ channelopa-thies and sinus node dysfunction. Prog Biophys Mol Biol 2008;98:171-8.

6. Drenth JP, Waxman sg. Mutations in sodium-channel gene SCN9A cause a spectrum of human genetic pain dis-orders. J clin invest 2007;117:3603-9.

7. Dabby r. Pain disorders and erythromelalgia caused by voltage-gated sodium channel mutations. curr Neurol Neurosci 2012;12:76-83.

8. levinson sr, luo sJ, Henry Ma. the role of sodium channels in chronic pain. Muscle Nerve 2011;46:155-65.

9. Maneksha Fr, Mirza H, Poppers PJ. complex regional pain syndrome (crPs) with resistance to local anesthetic block: a case report. J clin anesth 2000;12:67-71.

10. Panigel J, cook sP. a point mutation at F1737 of the hu-man Nav1.7 sodium channel decreases inhibition by local anesthetics. J Neurogenet 2011;25:134-9.

11. scholz a. Mechanisms of (local) anaesthetics on volt-age-gated sodium and other ion channels. Br J anaesth 2002;89:52-61.

12. Beckett Ha, gilmour ag. resistance to local analgesia--report of a case treated using 5% lignocaine solution. Br Dent J 1990;169:327-8.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: September 1, 2016. - Manuscript accepted: August 19, 2016. - Manuscript revised: July 28, 2016. - Manuscript received: June 15, 2016.(Cite this article as: Qu JZ, alston ta. When the nerve block is blocked by local resistance. Minerva anestesiol 2016;82:1029-31)


Observational studies

stroBe (strengthening the reporting of observational studies in epidemiology) state-ment deals with performing and reporting ob-servational studies.10, 11 accordingly, authors should include a flow diagram and a table that reports baseline demographic and clinical characteristics within each group. in observa-tional studies, baseline characteristics should be compared between groups in order to high-light unplanned differences. Unfortunately, the absence of statistically significant differences does not rule out the presence of confounders, nor supports the validity of univariate tests. in fact, comparisons between baseline values are often underpowered for demonstrating small imbalances that can invalidate the results of univariate statistical analysis. consequently, in observational studies with more than one group, comparisons of the outcomes should be preferably performed by multivariable models in order to adjust for the presence of potential-ly confounding factors.

observational studies include cohort, case-control, and cross-sectional studies. cohort studies are typically prospective, but can be

Reporting studies

the equator Web site offers links to all ma-jor guidelines issued for publication of

various types of studies.

Controlled clinical trials

the consort (consolidated standards of reporting trials) statement deals with ccts. in addition to the points considered in the “check-list” 1, 2 and its extension,3 authors should explic-itly declare whether they analyzed the intention-to-treat (itt) or per-protocol (PP) population.4, 5

A Consort flow diagram and a table report-ing baseline demographic and clinical char-acteristics in each group should be included in all manuscripts. in the table, many authors show the P-values about the comparison be-tween baseline characteristics.6, 7 However, this approach lacks of any statistical and scien-tific rationale and should be avoided. If found, statistically significant differences should not lead to the wrong conclusion that the groups belong to different populations, which is clear-ly illogical.8, 9 indeed, such differences are ex-plained with the probability of a type i error.

E D I T O R I A L

Basics to perform and present statistical analyses in scientific biomedical reports

Part 3Bruno M. cesana 1*, Franco cavaliere 2

1statistics and Biomathematics Unit, Department of Molecular and transactional Medicine, University of Brescia, Brescia, italy; 2Department of cardiovascular science, sacro cuore catholic University, rome, italy*corresponding author: Bruno M. cesana, statistics and Biomathematics Unit, Department of Molecular and transactional Medi-cine, University of Brescia, viale europa 11, 25143 Brescia, italy. e-mail: [email protected]

anno: 2016Mese: octobervolume: 82no: 10rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

lavoro: 11386-Mastitolo breve: Basics to perform and present statistical analyses in scientific biomedical reports: part 3primo autore: cesanapagine: 1032-5citazione: Minerva anestesiol 2016;82:1032-5

Minerva anestesiologica 2016 october;82(10):1032-5© 2016 eDiZioni Minerva MeDicaonline version at http://www.minervamedica.it

Basics to PerForM anD Present statistical analyses in scientiFic BioMeDical rePorts: Part 3 cesana


becomes 0.9583 (from 920/1000 divided by 960/1000), a not relevant value near to the no association value of 1.

Meta-analyses

Meta-analyses are usually aimed at new evi-dence from negative studies (without statisti-cal significant results) and/or to more precise estimates of effects (i.e. to narrower 95% cis). on the other hand, contrarily to primary stud-ies, systematic reviews and meta-analysis can hardly add relevant pieces of knowledge to a previous one published on the same subject shortly before. Hence, repeated meta-analyses adding only few trials without obtaining defi-nite results, should be regarded as unnecessary and unworthy of publication.

It is difficult to add any suggestion to the PrisMa (Preferred reporting items for sys-tematic reviews and Meta-analyses) state-ment,12 which consists of a 27-item checklist to which authors are referred. of particular importance is the inclusion of a PrisMa dia-gram, of a table that resumes data for each in-tervention group, and of forest plots to show effect estimates and confidence intervals. The use of i2 as a measure of consistency is rec-ommended. in order to facilitate the reader, authors are encouraged to provide some data as electronic supplements. For instance, crite-ria and data about possible biases within each study and across studies, as well as the charac-teristics of the studies included if their number is high (more than 20).

also retrospective; in both cases, two, usu-ally, cohorts of subjects (studies without a control group are rather case series studies) are followed during a lapse of time to search for a medical event or an outcome that occurs meanwhile. consequently, this type of obser-vational study often allows to formulate reli-able causality relationships by means of the relative risk (rr).

case-control studies start from a group of patients who presented a disease. a control group is successively formed with subjects, matched for the main factors possibly related to the disease. their purpose is to assess ret-rospectively the association between puta-tive risk factors and the outcome by means of the odds ratio (or), as it is obtained by the logistic regression analysis. Finally, cross-sectional studies assess the prevalence of a condition, mostly a disease, and the potential association with some considered factors. evidence for a relationship with the out-comes are much weaker than those assessed by case-control studies and especially, by co-hort studies.

or and rr are relative measures of associ-ation with a dichotomous outcome (improve-ment or not, success or not). the “odds” is the ratio between the probability of an outcome and the probability of its opposite; the risk of an outcome is the ratio between the number of subjects with the outcome and the num-ber of the observed subjects. let’s say that if 80 stroke are registered in 1000 smokers and 40 in 1000 nonsmokers, the odds of hav-ing a stroke is, therefore, 0.08696 (80/920) in smokers and 0.041 (40/960) in nonsmok-ers, whereas the risk of having a stroke is 0.080 (80/1000) in smokers and 0.040 (40/1000) in nonsmokers. Hence, the or of having a stroke in smokers vs. nonsmokers is 2.0869 (0.08696/0.041), while the rr is 2 (0.080/0.040), very similar to the or, as it is expected in the case of rare events. in-terestingly, or possesses an invariance prop-erty for which it is preferable to rr; indeed, the or of not having stroke for nonsmok-ers vs. smokers is 0.4791 (0.041/0.08696), the reciprocal of 2.0869. otherwise the rr,

Table I.—�Reporting sample size calculation (power analysis).

software versiontype “a priori”Statistical Significance Usually, α=0.05, Two-tailedPower Usually, 1-β=0.80effect size For means, indicatively:

– 0.20 considered small – 0.50 considered medium – 0.80 considered large

Usually, the variability value has to be reported

target difference Please justify the choicestandard deviation Please explain the source

cesana Basics to PerForM anD Present statistical analyses in scientiFic BioMeDical rePorts: Part 3


Sample size calculation (power analysis)

It is not ethically and scientifically accept-able that patients, animals and resources (time, money) are employed without an adequate probability of obtaining an evidence-based knowledge.13-17 therefore, a priori sample size calculations should be provided for all studies. In addition to the significance value alpha (usually α=0.05, two-tailed) and to the power (1-β, usually 0.80), authors should re-port the “effect size” 18 they aimed at, together with the considered variability value. the ef-fect size is the ratio between the minimum difference that authors deem clinically or bio-logically relevant in superiority trials (or the maximum difference not clinically or biologi-cally relevant in non-inferiority trials) 19, 20 and the phenomenon variability. the latter, in the case of a statistical test on the difference be-tween before and after a treatment corresponds to the standard deviation of the population of these differences. the statistical test for which the analysis was set and the software utilized should also be reported (tables i, ii).

in the case of only one group, the sample size is calculated for a test “against an expect-ed value” or for a required precision (95% ci width) of the main outcome estimate, even if the latter approach is not recommended.13-17 in order to be clinically sensible and plausible, the expected values or the required precision should be based on the literature available. in case of retrospective observational studies in which the sample size is predetermined, au-thors should report the differences between proportions (means) that can be demonstrated at a satisfactory power and significance level. alternatively, correlation and multivariate analysis can be used for sample size justifica-tion.23, 24 Finally, “a posteriori” power calcu-lations on already collected data is not statis-tically correct and, consequently, should be avoided.25

Sample size calculation is more difficult (even too much) and may require the involve-ment of a professional statistician in the case of repeated measurements on the same subject, multifactorial experimental designs (facto-

Table II.—�Check list for statistical analysis.

1. subsection on statistics in Methods section*a. type of study (examples: prospective cohort study, case-

control study, randomized clinical trial – rct -)b. adherence to recognized guidelines for that type of

studyc. list of all descriptive and inferential proceduresd. variable transformationse. Power analysisf. Missing datag. Significance levelh. statistical software (included the version)

2. results sectiona. suggested decimal places*b. Descriptive statistics*

i. Means (sD) orii. Medians (1st and 3rd quartiles; range)

c. correlation analysis#

i. Coefficient of determination, R2

d. regression analysis#

i. Intercept and regression coefficientwith 95% confidence intervals

ii. Coefficient of determination, R2

iii. Statistical significancee. Multiple regression analysis#

i. all the variables included in the analysis reported in Method section

ii. table with variablesin the last multivariable modeliii. Table with variables significant at the univariate

analysis (if more than four)f. Univariate survival analysis#

i. Kaplan-Meier product-limit methodii. graph of the survival functioniii. estimated hazard rateiv. number of patients at risk at selected times

g. agreement analysis#

i. Bland altman plotii. Mean difference between the two methods and

upper and lower limits of agreementiii. regression analysis between means and differences

h. controlled clinical trials§

i. intention-to-treat (itt) or per-protocol (PP) population

ii. table reporting baseline values (no statistical comparison)

iii. observational studies§

iv. Flow diagramv. table reporting baseline values (with statistical

comparison)vi. Meta analysis§

vii. PrisMa diagramviii. table resumingdata for each intervention groupix. Forest plot

i. Power analysis§

3. referencesa. references for non-standard statistical methods

*information contained in cesana BM et al.21; #information contai-ned in cesana BM et al.22; §information contained in this editorial.

Basics to PerForM anD Present statistical analyses in scientiFic BioMeDical rePorts: Part 3 cesana


ency of health research [internet]. Preferred reporting items for systematic reviews and Meta-analyses: the PrisMa statement available from:

13. cesana BM, reina g, Marubini e. sample size for test-ing a proportion in clinical trials: a “two step” proce-dure combining power and confidence interval expected width. the american statistician 2001;55:288-92.

14. cesana BM. sample size for testing and estimating the difference between two paired and unpaired proportions: a “two-step” procedure combining power and the prob-ability of obtaining a precise estimate. statistics in Medi-cine 2004;23:2359-73.

15. cesana BM, antonelli P. a new approach to sample size calculations for the power of testing and estimating population means of gaussian distributed variables. Bio-medical statistics and clinical epidemiology 2010;4:67-78.

16. cesana BM. Further insights in sample size calculation, Journal of Biopharmaceutical statistics 2013;23:937-9.

17. cesana BM, antonelli P. sample size calculations in clin-ical research should also be based on ethical principles. trials 2016;17:149 Doi 10.1186/s13063-016-1277-5.

18: effect size. Wikipedia [internet]. available from: https://en.wikipedia.org/wiki/effect_size [cited 2016, apr 2].

19. Points to consider on the choice of non-inferiority Mar-gin. london, 26 February 2004 cPMP/eWP/2158/99.

20. guideline on the choice of the non-inferiority margin. london, 27 July 2005. eMea/cPMP/eWP/2158/99.

21. cesana BM, cavaliere F. Basics to perform and present statistical analyses in scientific biomedical reports. Part 1. Minerva anestesiol 2016;82:822-6.

22. cesana BM, cavaliere F. Basics to perform and present statistical analyses in scientific biomedical reports. Part 2. Minerva anestesiol 2016;82:933-7.

23. Machin D, campbell M, tan sb, tan sh. sample size ta-bles for clinical studies. third edition. chichester, West sussex: Blackwell; 2009.

24. Julious sa. sample sizes for clinical trials. Boca raton, Fl: chapman &Hall/crc; 2009.

25. lenth rv. some Practical guidelines for effective sample size Determination. the american statistician 2001;187-93.

26. chow sc, shao J, Wang H. sample size calculations in clinical research. Boca raton, Fl: chapman & Hall/crc, inc.; 2008.

rial anova or ancova models), and mul-tivariable models (multiple linear regression, logistic regression, cox’s proportional hazard regression, etc.).26

References

1. schulz KF, altman Dg, Moher D, consort group. consort 2010 statement: updated guidelines for reporting parallel group randomized trials. BMJ 2010;340:c332.

2. schulz KF, altman Dg, Moher D, for the consort group. consort 2010 statement: Updated guidelines for reporting Parallel group randomized trials. ann in-tern Med 2010;152:726-32.

3. the consort Pro extension. JaMa 2013;309:814-22. 4. intention-to-treat analysis [internet]. available from

http://en.wikipedia.org/wiki/intention-to-treat_analysis [cited 2016, apr 2].

5. icH Harmonised tripartite guideline. statistical Prin-ciples For clinical trials e9 [internet]. available from http://www.ich.org/fileadmin/Public_Web_site/icH_Products/guidelines/efficacy/e9/step4/e9_guideline.pdf [cited 2016, apr 2].

6. senn s. testing for baseline balance in clinical trials. stat Med 1994;15:1715-26.

7. altman Dg. comparability of randomized groups. statis-tician 1985;34:125-36.

8. altman Dg. randomization: essential for reducing bias. Br Med J 1991;302:1481-2.

9. stroBe statement—checklist of items that should be included in reports of cross-sectional studies [internet]. available from http://www.strobe-statement.org/; [cited 2016, apr 2].

10. the Plos Medicine editors observational studies: get-ting clear about transparency Plos Med 11:e1001711.

11. equator network. enhancing the QUality and trans-parency of health research [internet]. available from: http://www.equator-network.org/reporting-guidelines/ [cited 2016, apr 2].

12. equator network. enhancing the QUality and transpar-

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: June 22, 2016. - Manuscript accepted: June 16, 2016. - Manuscript revised: May 18, 2016. - Manuscript received: april 12, 2016.(Cite this article as: Cesana BM, Cavaliere F. Basics to perform and present statistical analyses in scientific biomedical reports. Part 3. Minerva anestesiol 2016;82:1032-5)


in the last decades, the interest for High-Flow nasal cannula (HFnc) oxygen admin-istration as an alternative technique for niv or oxygenation has increased.2

The HFNC delivers oxygen flow rates from 40 to 60 liters per minute, improving gas ex-change through different mechanisms: by in-creasing co2 wash-out from the anatomical nasopharyngeal dead space (high Fio2 deliv-ered due to the absence of oxygen dilution), generating a low level of positive airway pres-sure (highly dependent of mouth-closing) en-suring the alveolar recruitment and increasing functional residual capacity.3

Moreover HFnc demonstrated an impor-tant feature of potential advantage over niv with possibility to bear oxygenation during laryngoscopy with a sort of “apneic oxygen-ation”.

HFnc could also be able to contrast desatu-ration related to o2 diffusion from the alveoli to the capillaries (that decreases alveolar pres-sure) by generating a flow from the pharynx to distal airways.

in this scenario, the review of ricard JD published in this issue of Minerva Aneste-siologica focuses on the strategies to prevent oxygen desaturation during endotracheal intu-bation of critically ill patients.4

endotracheal intubation is one of the most routinely performed invasive procedures,

but also frequently associated with morbidity and mortality.

severe complications can occur and hypox-emia is most often reported in case of difficult airway access, especially in high-risk oxygen desaturation, as well as obese, critically ill and pregnant patients or in icU where preoxygen-ation is much less efficient mainly due to pa-tients’ unstable cardiovascular or respiratory status.

in healthy subjects and in the anesthesia setting, preoxygenation allows the increase of apnea timing without appreciable arterial desaturation. nevertheless, it is known that preoxygenation promotes atelectasis, which is one of the mechanisms involved in the onset of hypoxemia during intubation.1

efforts to improve preoxygenation through optimized conventional facemask oxygen-ation and use of niv appear only marginally effective to prevent desaturation. niv pres-ents some limitations, like the compliance of the patient and the need of positive pressure removal to allow for laryngoscopy, thus avoid-ing any oxygen delivery during attempts.

E D I T O R I A L

The high nasal flow therapy for pre-oxygenation: a new strategy for conventional procedures

in intensive care settings?Pierpaolo terragni 1 *, andrea P. cossU 2

1anesthesiology and intensive care, University of sassari, sassari, italy; 2anesthesia and general intensive care Unit, aoU sassari, sassari, italy*corresponding author: Pierpaolo terragni, anesthesia and general intensive care Unit, Department of surgical sciences, University of sassari, viale san Pietro, 43 (07100), sassari, italy. e-mail: [email protected]


lavoro: 11516-Mastitolo breve: tHe HigH nasal FloW tHeraPY For Pre-oXYgenationprimo autore: terragnipagine: 1036-8citazione: Minerva anestesiol 2016;82:1036-8

comment on p. 1098.

Minerva anestesiologica 2016 october;82(10):1036-8© 2016 eDiZioni Minerva MeDicaonline version at http://www.minervamedica.it

tHe HigH nasal FloW tHeraPY For Pre-oXYgenation terragni


tion during apnea, in simulated endotracheal intubation, increased the time until patients became severely hypoxic but did so only when the shunt fraction was below about 25% on a PeeP of 5 cm H2o.7

on this topic some authors, using HFnc in 20 infants undergoing endotracheal intuba-tion, found that the benefit remarkably differed between those with healthy or injured lungs: during the trial, if desaturation episodes devel-oped, the mouth was gently closed to increase pharyngeal pressure.8

HFnc technique represents a real attractive topic and its application for preventing hypox-emia during intubation in critically ill patients is acquiring interest, as demonstrated by the grow-ing number of clinical trials on this argument.9, 10

as reported by the review of ricard JD,4 a growing body of evidence suggests that HFnc oxygen therapy is an innovative and ef-fective technique to improve oxygenation and prevent desaturation with different underlying diseases.

However, there is no therapy that is efficient in every patient and in every type of respira-tory failure.

Despite promising preliminary data, evi-dence underscore the need for rcts to test the effects of HFnc before and during endotra-cheal intubation in patients stratified according to lung disease.

References

1. Hedenstierna g, edmark l. Mechanisms of atelectasis in the perioperative period. Best Pract res clin anaesthe-siol 2010;24:157-69.

2. Ricard JD. High flow nasal oxygen in acute respiratory failure. Minerva anestesiol 2012;78:836-41.

3. chanques g, riboulet F, Molinari n, carr J, Jung B, Prades a, et al. Comparison of three high flow oxygen therapy delivery devices: a clinical physiological cross-over study. Minerva anestesiol 2013;79:1344-55.

4. ricard JD. Hazards of intubation in the icU: role of nasal high flow oxygen therapy for preoxygenation and apneic oxygenation to prevent desaturation. Minerva anestesiol 2016;82:1098-106.

5. Mort tc, Waberski BH, clive J. extending the pre-oxygenation period from 4 to 8 mins in critically ill pa-tients undergoing emergency intubation. crit care Med 2009;37:68-71.

6. Frizzola M, Miller tl, rodriguez Me, Zhu Y, rojas J, Hesek a, et al. High-flow nasal cannula: impact on oxy-genation and ventilation in an acute lung injury model. Pediatric Pulmonology 2011;46:67-74.

the author describes how several factors can contribute to increase the risks of life-treating hypoxemia during intubation maneu-vers in emergency situations.

conventional preoxygenation has a limited efficacy in ICU patients for several reasons such as hemodynamic instability, ventilation/perfusion mismatch, obesity etc.5

author suggests an algorithm where high flow nasal oxygen should be taken into account in case of moderate or severe hypoxemia and added to niv during laryngoscopy for very se-vere hypoxemic patients.

During orotracheal intubation the decrease of alveolar pressure, due to oxygen absorption in the apnea phase, generates a pressure gradi-ent between alveolar space and pharynx pro-moting a gas flow from upper to distal airways in absence of respiratory movement.

In this case, the efficacy of HFNC could be related to the optimal conditioning of inspired gases with high Fio2 delivered (due to the ab-sence of oxygen dilution) coupled to the pha-ryngeal dead-space washout, rather than the amount of positive pressure generated during apneic oxygenation (presenting, during intuba-tion maneuver, a neurological impairment of the patient).

as reported by Frizzola et al. in an experimen-tal model on the effects of HFnc, with respect to gas exchange, the impact of increasing flow on ventilation and oxygenation occurs regard-less of the only tracheal pressure generation.6 in the injured lung model, by increasing cPaP pressure, neither Paco2 nor Pao2 demonstrated progressive changes; however, with HFnc un-der leak conditions (somewhat as if it was com-parable to the open mouth during laryngoscopy) Paco2 and Pao2 improved in a flow dependent response, so that saturation relationships where consistent, in the experimental model, with na-sopharyngeal dead space washout as demon-strated in several clinical conditions.

However the benefits of HFNC in patients undergoing endotracheal intubation could be not constant between those with healthy or in-jured lungs.

in experimental conditions, engström et al. showed that pharyngeal oxygen administra-

terragni tHe HigH nasal FloW tHeraPY For Pre-oXYgenation


n, egreteau PY, et al. High-flow nasal cannula oxygen during endotracheal intubation in hypoxemic patients: a randomized controlled clinical trial. intensive care Med 2015;41:1538-48.

10. simon M, Wachs c, Braune s, de Heer g, Frings D, Kluge s. High-Flow nasal cannula oxygen versus Bag-valve-Mask for Preoxygenation Before intubation in subjects With Hypoxemic respiratory Failure. respir care 2016 Jun 7. pii: respcare.04413 [epub ahead of print].

7. engstrom J, Hedenstierna g, larsson a. Pharyngeal oxy-gen administration increases the time to serious desatura-tion at intubation in acute lung injury: an experimental study. crit care 2010;14:r93.

8. Papoff P, luciani s, Barbara c, caresta e, cicchetti r. High-Flow nasal cannula to Prevent Desaturation in en-dotracheal intubation: a Word of caution. crit care Med 2015;43:e327-8.

9. vourc’h M, asfar P, volteau c, Bachoumas K, clavieras

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: July 22, 2016. - Manuscript accepted: July 19, 2016. - Manuscript received: June 29, 2016.(Cite this article as: Terragni �, Cossu A�. The high nasal fl ow therapy for pre-oxygenation: a new strategy for conventional proce-Terragni �, Cossu A�. The high nasal flow therapy for pre-oxygenation: a new strategy for conventional proce-dures in intensive care settings? Minerva anestesiol 2016;82:1036-8)


and increased mortality.2, 3 colloids in contrast are retained at the functioning barrier and remain intravascularly. this different behaviour means they are not indicated for fluid losses but might be advantageous for intravascular losses. Many trials exist showing that the perioperative use of colloids improves fluid balance and reduces ede-ma formation.4, 5 Moreover, using a goal-directed approach during surgery in hypovolemic patients iso-oncotic hydroxyethyl starch (Hes) has shown to achieve a faster and longer-lasting haemody-namic stabilisation, less complication rates, im-proved lung function, faster gut recovery, shorter hospital stay and shorter duration of ventilation.6, 7 Using dynamic preload parameters several trials have also shown that this colloid significantly in-creased stroke volume, cardiac output, corrected flow time as well as an improved microcircula-tory blood flow and oxygen delivery to organs in comparison to crystalloids.4, 8, 9 some argue that better hemodynamics and fluid balances are not clinically relevant as despite those improvements some trials show no outcome differences. How-ever, a closer look at the study protocols can be useful. two recent perioperative double-blind rcts comparing Hes with crystalloids had to switch to “rescue colloids” when the study drug dose limitation of 50 ml/kg was achieved. as a consequence, the patients in the crystalloid group received considerable amounts of colloids (either 1.5 l plasma 4 or 700 ml gelatin 10). Does that show that colloids and crystalloids have similar

one of the anaesthesiological hot topics of the past years has been the colloid vs. crys-

talloid controversy. this meanwhile emotional and irrational debate has moved from the sci-entific area to the yellow press journalism. The colloid objectors are leading a crusade against these products trying to convert all non-believ-ers, thereby ignoring physiological principles and twisting evidence and facts.

in this issue of Minerva Anestesiologica, Pis-ano et al. demonstrate this one-sided view quite nicely.1 crystalloids are made out to be the one-size-fits-all drug, being ideal to replace any kind of fluid losses including acute bleeding or in acute shock. crystalloids consist of water and electro-lytes and are evenly distributed over the entire ex-tracellular space (divided into 80% interstitial and 20% intravascular space). Whereas losses from this compartment, e.g. urine output and insensible perspiration, are ideally treated with crystalloids, replacing losses from the intravascular compart-ment, e.g. blood losses, does not make sense. How can a drug be ideal when only 20% remain in the target compartment and 80% is shifted into tissue where it is ineffective and not needed? Moreover, this fluid overload presents as interstitial oedema, impairs microcirculation, organ perfusion and causes subsequent complications such as pulmo-nary and gut oedema, pneumonia, prolonged ven-tilation, prolonged gut recovery, abdominal com-partment syndrome, impaired wound healing,

E D I T O R I A L

colloidophobiaJavier riPollÉs MelcHor 1*, Dietmar Fries 2, Daniel cHaPPell 3

1Department of anesthesia and critical care, infanta leonor University Hospital, Madrid, spain; 2Department for general and surgical intensive care Medicine, Medical University innsbruck, innsbruck, austria; 3Department of anaesthesiology, University Hospital of Munich (lMU), Munich, germany*corresponding author: Javier ripollés Melchor, goya 111, 2 D, 28009, Madrid, spain. [email protected]


lavoro: 11415-Mastitolo breve: colloidophobiaprimo autore: riPollÉs MelcHorpagine: 1039-42citazione: Minerva anestesiol 2016;82:1039-42


comment on p. 1107.

riPollÉs MelcHor colloiDoPHobia


or de novo disruption, the effects of all colloids and crystalloids result in very similar effects on morbidity and mortality.16 as with every drug, side-effects are dependent on the dose, the con-text, the indication and the product. old starch preparations with high molecular weight have been frequently associated with kidney dys-functions and coagulation disorders. two meta-analyses in surgical patients 17, 18 have not shown these effects with 6% Hes 130, the only Hes solution that should be used today. addition-ally, the recently published german guidelines on intravascular volume therapy (approved by 14 National societies) stated that there is “no indication of renal insufficiency associated with the peri-interventional use of Hes 130, gelatine or albumin”. both above-mentioned sepsis trials have been criticised for their methodology and interpretation.20 the ViseP study compared a hyperoncotic 10% Hes 200 solution (with 40% of patients receiving more than the maximally allowed amount) with a balanced crystalloids solution.13 in the 65 study, large amounts of iso-oncotic Hes 130 were administered during several days following successful hemodynamic stabilisation 14 according to the criteria of the surviving sepsis campaign. in both trials, col-loids were part of the initially administered intra-vascular fluids in both study groups. At the time of randomization 12-24 hours after diagnosis of sepsis, most patients were hemodynamically stabilized. thus, the continued administration of Hes in the Hes group was highly question-able”.16, 20 at the end of the trials over 90% of their respective crystalloid group had received colloids during their hospital stay. again — if pure crystalloids are so successful why use col-loids in the vast majority of patients? And why not discuss this aspect objectively and critically? if there are signs of adverse drug effect, it can be because of the drug itself or misuse of it. the re-cent guidelines took up and confirmed these crit-icisms: “these studies offer contradictory results and exhibit methodological deficiencies”.19 the third critical care trial is the currently under-fire cHest trial.21 their Hes group had a better kidney function according to the riFle score, but also a 1% increased rate of renal replacement therapy (RRT) in the unadjusted analysis. This

effects and that only giving crystalloids is equally effective? or rather that crystalloids alone are not able to achieve sufficient haemodynamic stabili-sation? the only perioperative trial mentioned by Pisano et al.1 (all above mentioned aspects and trials were ignored) was the trial from bayer et al. comparing Hes, gelatine and crystalloids in cardiac surgery.11 their Hes group experienced a better haemodynamic stability, less fluid over-load, better organ function and shorter length of mechanical ventilation — the authors erroneous-ly interpreted the findings as showing HES to be “not more effective” than crystalloids. their gela-tine group actually received no gelatine in the first 3 days. on the other hand the “only crystalloid” group received 1,000 ml of Hes as priming so-lution and during the trial 126 ml/kg “non-crys-talloids” (e.g. an 80 kg patient received 10.080 ml non-crystalloids), unfortunately it remains unclear what these fluids were. The same group stated to see no difference in resuscitating patients in septic shock when administering Hes or “only crystalloids”.12 only crystalloids meant infusing 1.500 (interquartile range [iQr] 1.000-2.000) ml fresh frozen plasma and 560 (320-1120) ml 20% human albumin, which corresponds to 2.240 (1.280-4.480) ml of isooncotic 5% albumin. if a sole crystalloid therapy is so effective why do the authors always use colloids too? but more impor-tantly why ignore this fact and not lead an open and critical discussion?

Pisano et al. main criticism on Hes is based on 3 trials — those that caused the official au-thorities to get involved in 2013. two of these trials included patients in septic shock.13, 14 this patient population completely differs from the perioperative one in that those patients suffer from an impaired permeability barrier with sub-sequent capillary leak.3 the main causes of this barrier impairment are disruptions of glycocalyx and endothelial cell junctions. Glycocalyx de-gradation is known to be caused by, e.g., sep-sis, trauma, ischemia, cytokines and iatrogenic hypervolaemia.15 in septic shock this disruption causes colloids to be shifted to a large part into tissue, losing their advantage of high volume effects and increasing the risk of side effects. this could explain why, in the critically ill with either chronic alteration of the glycocalyx and/

colloiDoPHobia riPollÉs MelcHor


Pisano et al. draw their negative conclusions on gelatine without presenting any confirm-ing data. gelatins have been successfully used for over 100 years providing a large amount of clinical experience with its safety profile. Over time the formulation has been improved several times from oxypolygelatine (1951) to urea-cross linked gelatine (1958), succinylated gelatine (1962) and succinylated gelatine dissolved in balanced solutions (2013). the gelatin mol-ecule is broken down into smaller fragments and eliminated by renal enzyme systems as soon as falling below the renal threshold. according to current knowledge gelatine and its derivatives are not stored in any tissue or organ. Despite that only few trials on the use of gelatine exist, no organ complications due to tissue storage or ac-cumulation have been reported. a meta-analysis including 30 randomized controlled trials (2709 patients) did not detect any differences with re-gards to renal failure, blood loss or mortality comparing gelatine to crystalloids.27 above that, there seems to be no clinical relevant interac-tion and impairment of the coagulation system so that gelatine promises to be a good option in critically ill patients. this is also part of the ger-man guidelines on volume therapy, which state that albumin, gelatine and Hes 130 are equiva-lent and recommend the use of gelatine in acute hypovolaemia in icU patients if crystalloids alone are inadequate,19 which is exactly what the above-mentioned trials have shown. Pisano and colleagues recommend not to use gelatins and blame them of being old. this undifferentiated statement ignores decades of clinical experience, facts on different gelatines over time and recom-mendations from high quality guidelines.28

James Hogan wrote in JaMa in 1915 “re-suscitation with colloid is more effective than saline for hypovolemic shock, but insufficient to treat toxemic shock, despite the initial effects of colloid resuscitation on blood pressure.”29 over 100 years later authors trying to prove him wrong are failing. sometimes old things can be quite useful. in order to solve the colloid vs. crystalloid debate an open, objective, rea-sonable and unbiased discussion on a scientific basis is required. but sometimes ignorance is simply bliss.

difference in RRT did not exist in the adjusted analysis, published in the supplement, however the improved kidney function remained. in this general icU collective the better kidney func-tion might confirm the findings of James et al. who showed an improved kidney function af-ter resuscitation with Hes vs. crystalloids in a double-blind rct in trauma patients.22 Most likely the reduction of prerenal failure due to the faster stabilisation might be the reason for this. in order to evaluate this aspect and to understand the cHest findings also in the light that 36% of their patients suffered from acute renal failure at study onset (a well-known contraindication for Hes and from our point of view a reason to exclude them from randomization) the authors were asked to share their raw data in 2013 by the sponsoring company. this was denied. the only raw data released were the adverse events – which the authors have meanwhile admitted to be wrong: “caused by a typographical tran-scriptional error”.23 in March 2016 the bMJ published an article criticising both the NeJM and the authors for not releasing their data, not publishing an erratum or even the original proto-col.24 the bMJ even raised the question if those who control the data actually control the mes-sage and demanded a release of the raw data. a few days after this article the adverse events in chest were updated by the NeJM, however after changing the initial statistical analysis plan 25 and 24 patients missing in the new table. according to Priebe, if the intended itt analysis had been carried out there would be no difference of ad-verse events.23 in the light of these misleading interpretations, missing patients and methologi-cal deficiencies we join the BMJ in asking the authors of the cHest study to make public the raw data and have them re-analyzed by outside reviewers. a further point why it is important to understand these trials is the fact that the only trial analysing initial volume resuscitation dur-ing shock is the trial from annane et al.26 they randomized patients in acute hypovolemic shock directly at arrival in icU. the infusion of 2000 mL colloids in this phase significantly improved 90d mortality in comparison to 3000 ml crys-talloids, showing that a rational indication-based use of colloids may have clear benefits.

riPollÉs MelcHor colloiDoPHobia


130/0.42 versus ringer’s acetate in severe sepsis. N engl J Med 2012;367:124–34.

15. becker bF, Jacob M, leipert s, salmon aH, chappell D. Degradation of the endothelial glycocalyx in clinical set-tings: searching for the sheddases. br J clin Pharmacol 2015;80:389-402.

16. De robertis e, afshari a, longrois D. the quest for the holy volume therapy. eur J anaesthesiol 2016;33:483-7.

17. Van Der linden P, James M, Mythen M, Weiskopf rb. safety of modern starches used during surgery. anesth analg 2013;116:35-48.

18. Gillies MA, Habicher M, Jhanji S, Sander M, Mythen M, Hamilton M, Pearse rM. incidence of postoperative death and acute kidney injury associated with i.v. 6% hydroxyethyl starch use: systematic review and meta-analysis. br J anaesth 2014;112:25-34.

19. Marx g, schindler aW, Mosch c, albers J, bauer M, gnass i, et al. intravascular volume therapy in adults: guidelines from the Association of the Scientific Medical Societies in germany. eur J anaesthesiol 2016;33:488-521.

20. chappell D, Jacob M. Hydroxyethyl starch - the impor-tance of being earnest. scand J trauma resusc emerg Med 2013;21:61.

21. Myburgh Ja, Finfer s, bellomo r, billot l, cass a, gat-tas D, et al. Hydroxyethyl starch or saline for fluid resus-citation in intensive care. N engl J Med 2012;367:1901-11.

22. James MF, Michell Wl, Joubert ia, Nicol aJ, Navsaria PH, gillespie rs. resuscitation with hydroxyethyl starch improves renal function and lactate clearance in penetrat-ing trauma in a randomized controlled study: the First trial (Fluids in resuscitation of severe trauma). br J anaesth 2011;107:693-702.

23. Priebe HJ. response to Data too important to share: do those who control the data control the message? bMJ 2016;352:i571.

24. Doshi P. Data too important to share: do those who con-trol the data control the message? bMJ 2016;352:i1027.

25. Doshi P. Update: New england Journal of Medicine pub-lishes correction to 2012 cHest trial of hydroxyethyl starch versus colloids. bMJ 2016 Mar 21;352:i1571.

26. annane D, siami s, Jaber s, Martin c, elatrous s, De-clère aD, et al. Effects of fluid resuscitation with col-loids vs crystalloids on mortality in critically ill patients presenting with hypovolemic shock: the cristal rand-omized trial. JaMa 2013;310:1809-17.

27. Saw MM, Chandler B, Ho KM. Benefits and risks of us-ing gelatin solution as a plasma expander for periopera-tive and critically ill patients: a meta-analysis. anaesth intensive care 2012;40:17-32.

28. basora M, colomina MJ, Moral V, asuero de lis Ms, boix e, Jover Jl, et al. clinical practice guide for the choice of perioperative volume-restoring fluid in adult patients undergoing non-cardiac surgery.. rev esp anes-tesiol reanim 2016;63:29-47.

29. Hogan JJ. the intravenous use of colloidal (gelatin) solu-tions in shock. JaMa 1915;lXiV:721-726.

References

1. Pisano a, landoni g, bellomo r. the risk of infusing gelatin? Die-hard misconceptions and forgotten (or ig-nored) truths. Minerva anestesiol 2016;82:1107-14.

2. besen ba, gobatto al, Melro lM, Maciel at, Park M. Fluid and electrolyte overload in critically ill patients: an overview. World J crit care Med 2015;4:116-29.

3. chappell D, Jacob M, Hofmann-Kiefer K, conzen P, Rehm M. A rational approach to perioperative fluid ther-apy. anesthesiology 2008;109:723-40

4. Feldheiser a, Pavlova V, bonomo t, Jones a, Fotopoulou c, sehouli J, et al. balanced crystalloid compared with balanced colloid solution using a goal-directed haemody-namic algorithm. br J anaesth 2013;110:231-40.

5. Yates Dr, Davies sJ, Milner He, Wilson rJ. crystalloid or colloid for goal-directed fluid therapy in colorectal sur-gery. br J anaesth 2014;112:281-9.

6. benes J, chytra i, altmann P, Hluchy M, Kasal e, svitak r, et al. Intraoperative fluid optimization using stroke volume variation in high risk surgical patients: results of prospective randomized study. crit care 2010;14:r118.

7. goepfert Ms, richter HP, Zu eulenburg c, gruetzmacher J, Rafflenbeul E, Roeher K, et al. individually optimized hemodynamic therapy reduces complications and length of stay in the intensive care unit: a prospective, rand-omized controlled trial. anesthesiology 2013;119:824-36.

8. Kanda H, Hirasaki Y, iida t, Kanao M, toyama Y, Kuni-sawa t, et al. Effect of fluid loading with normal saline and 6% hydroxyethyl starch on stroke volume variability and left ventricular volume. int J gen Med 2015;8:319-24.

9. Kimberger o, arnberger M, brandt s, Plock J, sigurds-son gH, Kurz a, et al. goal-directed colloid adminis-tration improves the microcirculation of healthy and perianastomotic colon. anesthesiology 2009;110:496-504.

10. Yates Dr, Davies sJ, Milner He, Wilson rJ. crystalloid or colloid for goal-directed fluid therapy in colorectal sur-gery. br J anaesth 2014;112:281-9.

11. bayer o, schwarzkopf D, Doenst t, cook D, Kabisch b, schelenz c, et al. Perioperative fluid therapy with tetrastarch and gelatin in cardiac surgery - a prospective sequential analysis. crit care Med 2013;41:2532-42.

12. bayer o, reinhart K, Kohl M, Kabisch b, Marshall J, sakr Y, et al. Effects of fluid resuscitation with synthetic colloids or crystalloids alone on shock reversal, fluid balance, and patient outcomes in patients with severe sepsis: a prospective sequential analysis. crit care Med 2012;40:2543-51.

13. brunkhorst FM, engel c, bloos F, Meier-Hellmann a, ragaller M, Weiler N, et al. intensive insulin therapy and pentastarch resuscitation in severe sepsis. N engl J Med 2008;358:125-39.

14. Perner a, Haase N, guttormsen ab, tenhunen J, Kle-menzson g, aneman a, et al. Hydroxyethyl starch

Conflicts of interest.—JrM: Honoraria for lectures and travel expenses from Deltex Medical, edwards lifesciences and Fresenius Kabi. DF: Honoraria for consulting, lectures fee and sponsoring for academia studies from astra Zeneca, aoP orphan, baxter, bayer, bbraun, biotest, csl behring, Delta select, Dade behring, edwards, Fresenius Kabi, glaxo, Haemoscope, Hemogem, lilly, LFB, Mitsubishi Pharma, NovoNordisk, Octapharm, Pfizer, and Tem-Innovation. DC: Honoraria for lectures and academic studies from bbraun, Fresenius Kabi, grifols, and lFb biomedikaments.Article first published online: June 28, 2016. - Manuscript accepted: June 27, 2016. - Manuscript revised: June 15, 2015. - Manuscript received: May 3, 2016.(Cite this article as: ripollés Melchor J, Fries D, chappell D. colloidophobia. Minerva anestesiol 2016;82:1039-42)


tricle (lV) functions with echocardiography in a small subset of patients with acute respirato-ry failure, an increasing number of papers have been published on this topic, but the available

since the first report in 1985 by Jardin et al. 1 describing the right (rV) and left ven-

O R I G I N A L A R T I C L E

right ventricle dilation as a prognostic factor in refractory acute respiratory

distress syndrome requiring veno-venous extracorporeal membrane oxygenation

chiara laZZeri 1 *, giovanni ciaNcHi 2, Manuela BoNiZZoli 2, stefano BataccHi 2, Paolo tereNZi 2, Pasquale BerNarDo 1, serafina ValeNte 1, gian Franco geNsiNi 1, 3, adriano Peris 2

1intensive cardiac care Unit, Heart and Vessels Department, University Hospital of careggi, Florence, italy; 2intensive care Unit and regional ecMo referral centre, emergency Department, University Hospital of careggi, Florence, italy; 3Department of experimental and clinical Medicine, University Hospital of careggi, Fondazione Don carlo gnocchi irccs, Florence, italy*Corresponding author: Chiara Lazzeri, Intensive Cardiac Care Unit, Heart and Vessel Department, Viale Morgagni 85, 50134 Flor-ence, italy. e-mail: [email protected]


Lavoro: 10934-MAStitolo breve: rV DilatatioN as a ProgNostic Factor iN arDsprimo autore: laZZeripagine: 1043-9citazione: Minerva Anestesiol 2016;82:1043-9

a B s t r a c tBacKgroUND: the aim of this study was to assess the incidence and prognostic role of echocardiographic abnormali-ties in consecutive patients with refractory acute respiratory distress syndrome (arDs) before veno-venous extracorpor-eal membrane oxygenation (VV-ecMo).MetHoDs: in this study 74 consecutive patients with refractory arDs underwent echocardiography (transthoracic, transesophageal or both, according to the best acoustic window). Baseline characteristics were collected for all patients and the simplified acute physiology score was calculated. At echocardiography the following parameters were consid-ered: left ventricle (lV) ejection fraction, right ventricle (rV) size and function (by means of tricuspid annular plane excursion [taPse]) and systolic pulmonary arterial pressure.RESULTS: At echocardiography, 25 patients showed normal findings (33.8%), 32 patients exhibited isolated pulmo-nary hypertension (43.2%) and the remaining 17 patients showed RV dilation and pulmonary hypertension (23%). A reduced LVEF (<50%) was observed in 14 patients (18.9%), while RV dysfunction (as indicated by TAPSE<16 mm) was documented in 21 patients (28.4%). The in-Intensive Care Unit [ICU] mortality rate was 41.8%. At stepwise regression analysis the following variables were independent predictor for in-icU mortality (when adjusted for taPse<16 mm): RV end diastolic area/LV end diastolic area (OR 0.21, 95%CI 0.062-0.709, P=0.012), Body Mass Index (BMI) (OR 0.87, 95%CI 0.802-0.958, P=0.004)coNclUsioNs: in consecutive patients with refractory arDs, echocardiographic alterations were common, mainly represented by systolic pulmonary hypertension associated or not with rV dilatation. Moreover, rV dilatation and BMi were independent predictors of in-ICU mortality. On clinical grounds, our findings strongly suggest that echocardiogra-phy helps to risk stratifying patients with refractory arDs requiring VV-ecMo.(Cite this article as: lazzeri c, cianchi g, Bonizzoli M, Batacchi s, terenzi P, Bernardo P, et al. right ventricle dilation as a prognostic factor in refractory acute respiratory distress syndrome requiring veno-venous extracorporeal membrane oxygen-ation. Minerva Anestesiol 2016;82:1043-9)Key words: echocardiography - respiratory distress syndrome, adult - Heart ventricles.

comment in p. 1023.

Minerva Anestesiologica 2016 October;82(10):1043-9© 2016 eDiZioNi MiNerVa MeDicaonline version at http://www.minervamedica.it

laZZeri rV DilatatioN as a ProgNostic Factor iN arDs


of ecMo cannulas by transesophageal/trans-thoracic ultrasonography.7, 9, 11, 12

according to our protocol 9, 11, 12, 14 the echocardiographic examination is transtho-racic (tte), transesophageal (tee) or both, according to the best acoustic window (es-aote Mylab™30gold cardiovascular, esaote s.p.a, genoa, italy).15 the lV ejection frac-tion (lVeF) was estimated by eyeball exami-nation on short-axis views.16 lV systolic dys-function was defined as LV ejection fraction less than 45%.17

the right ventricle (rV) size was assessed by the rV end-diastolic area (eDa) (four-chamber view) and the ratio between eDas of the right and lVs was calculated (rVeDa/LVEDA). This ratio classifies RV function as normal (<0.6), moderately altered (0.6-0.8), severely altered (>0.8).18

systolic pulmonary artery pressure (sPaP) was obtained using the simplified Bernouilli’s equation: 4 • (Vmax tricuspid regurgitation)2 + central venous pressure (cVP). to reduce the lack of precision of cVP estimation based on the size of the inferior vena cava, cVP was invasive-ly measured through central venous catheters.5, 15

tricuspid annular plane excursion (taPse) 5 was also measured, as the difference of dis-placement during diastole and systole.5, 17 a taPse<16 mm is known to indicate rV dys-function.5, 15, 19

the requirement of ecMo implantation was decided on the basis of the italian Ministry of Health criteria, as previously described.11, 12

According to echocardiographic findings, the following subgroups of patients were con-sidered:17

a) group 1: Normal rV dimensions and function, normal values of sPaP;

b) group 2: normal rV dimensions and func-tion, values of sPAP≥40 mmHg;

c) group 3: rV dilatation and values of sPAP≥40 mmHg

Mortality during icU stay was the outcome.all participants (or their kins) signed a writ-

ten inform consent for storing their clinical data. the study is a retrospective analysis of data and the study design was approved by our institutional Board.

evidence mainly concerns the prevalence and the prognostic impact of acute cor pulmonale (acP) in acute respiratory distress syndrome (arDs).2-5

Veno-venous extracorporeal membrane ox-ygenation (VV-ecMo) is a well-established therapy in patients with arDs unresponsive to conventional therapy.6-8 However, to date, there is a paucity of data on cardiac function assessed by echocardiography in these pa-tients, since this technique is mainly used in combination with other data for selecting the type of ecMo, as a visual guide and monitor the cannulation and in the early detection of complications during the procedure of ecMo implantation.9

the present investigation was aimed at as-sessing the incidence and prognostic role of echocardiographic abnormalities in 74 con-secutive patients with refractory arDs before VV-ecMo.

Materials and methods

From oct 10 2009 to December 31 2013, seventy four patients with refractory arDs 10 requiring VV-ecMo were consecutively ad-mitted to our intensive care Unit, which is an ecMo referral center. according to our protocol,9, 11, 12 an echocardiographic exam is performed just before ecMo implantation (echo ecMo Florence registry). Data were prospective recorded and retrospective ana-lyzed.

Baseline characteristics were collected for all patients and the Simplified Acute Physiol-ogy score (saPs ii) score was calculated.13 all patients were mechanically ventilated. the ratio Pao2/Fio2 was considered a criterion for arDs severity.

echocardiography is performed routinely before ecMo implantation at our intensive care Unit (icU) by the cardiologist (cl and PB) who is part of our ecMo team which also includes an intensivist, a cardiac surgeon and a perfusionist, all trained on ecMo technique and management. The cardiologist’s main task is to evaluate cardiac function in the pre-ec-Mo phase and guides the correct positioning

rV DilatatioN as a ProgNostic Factor iN arDs laZZeri


in-icU mortality. Hosmer-lemeshow test as-sessed the calibration of the logistic model and a c-statistic was used to test its goodness-of fit. We assessed intraobserver variability with an intraclass correlation test and interobserver variability with a κ test. A κ up than 0.8 was considered as excellent. the intraclass correla-tion coefficients are used to assess agreement of quantitative measurements in the sense of consistency which can be defined as the agree-ment of two quantitative measurements in set-tings where neither one is assumed “correct” and therefore handles questions of intra- as well as interobserver reproducibility of mea-surement scales.20

Results

intraobserver variability was quite good with an intraclass correlation of 0.89 (95% confidence interval, 0.85-0.93) and interob-server variability was also excellent with a κ of 0.96 (95% confidence interval, 0.94-0.97).

our population comprises mainly males (54/74, 72.9%) (Table I) and ARDS was mod-

Statistical analysis

statistical analysis has been conducted with SPSS 13.0 for Windows software (SPSS inc, chicago, il, Usa). a two-tailed P-value <0.05 was considered statistically significant. categorical variables are reported as frequen-cies and percentages; continuous variables are reported as mean±standard deviation (sD). For continuous variables, between-groups comparisons have been performed with stu-dent’s t-test or aNoVa (followed by Bonfer-roni post-tests if overall P was significant) or by means of Kruskal-Wallis H Test. Categori-cal variables have been compared with χ2.

Univariate analysis (χ2or Fisher’s Exact Test for categorical data; Student’s t-test or Mann-Whitney U Test for continuous data) was used to identify candidate variables (age, gender, Body Mass index [BMi], po2, pco2, pH, lVeF, rVeDa/lVeDa ratio, taPse, sPaP) for multivariate analysis using a significant threshold.

Backward stepwise logistic regression was performed in order to identify predictors of

Table I.—�Clinical characteristics and echocardiographic findings.all

patients icU patients Patients transferred from peripheral hospitals P

Number 74 42 32age (yrs) 51.5 ±15.1 52.7±16.7 49.9±12.9 NssaPs ii 44.1 ±19.2 41.7±18.9 46.5±15.4 NssoFa 10.7±2. 10.3±3.2 10.2±2.3 NsMales/females 54/74 32/10 22/10/15 NsBMi 27.9 ±8.5 27.9±8.5 29.5±8.8 NsecMo duration (days) 13.5±11.2 14.5±14.2 12.2±6.6 NsMechanical ventilation (days) 19.9±14.3 20.0±15.9 21.5±13.2 NsicU los (days) 23.9±17.1 23.6±19.6 22.7±14.7 NsVasopressors 37 20 17Norepinephrine only 29 15 14 NsNorenephrine and dobutamine 6 4 2Dobutamine only 2 1 1Blood gas analysis

Pao2/Fio2 ratio<300 9 4 5 Ns<200 21 14 7<100 44 24 20

pH 7.36±0.12 7.35±0.10 7.36±0.12 Nspco2 49.8±26.3 50.9±23.0 46.5±20.3 Nspo2 75.2±23.9 87.6±40.7 82.5±34.4 Nslactate (mg/dl) 2.2±1.9 3.6±3.9 3.2±4.7 Ns

ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit: SAPS II: simplified acute physiology score; BMI: Body Mass Index; los: length of stay.



and the remaining 17 patients showed rV di-lation and pulmonary hypertension (group 3, 23%). A reduced LVEF (<50%) was observed in 14 patients (18.9%), while RV dysfunction (as indicated by taPse<16 mm) was docu-mented in 21 patients (28.4%) (Table II). One patient exhibited acute cor pulmonale and was included in group 3. in the overall population, the in-ICU mortality rate was 41.8%. No dif-ference was observed between icU patients and those who were transferred from periph-eral centers.

erate-to severe in most cases (87.8%) and due to H1N1 infection in 23 patients, H3n2 in 2 pa-tients, viral pneumonia in 8 patients and bacte-rial pneumonia in the remaining 43 patients. echocardiography was transthoracic in 26 pa-tients (35.1%), transesophageal in 29 patients and both (transthoracic and transesophageal) in the remaining 29 patients (29.2%). Accord-ing to echocardiography examination (tables II, III), 25 patients showed normal findings (group 1, 33.8%), 32 patients exhibited isolat-ed pulmonary hypertension (group 2, 43.2%)

Table II.—�Echocardiographic findings.

all patients icU patients Patients transferred from peripheral hospitals P

Echocardiographic findingsgroup 1 25 (33.8%) 18 7 Nsgroup 2 32 (43.2%) 17 15group 3 17 (23%) 7 10

tte 26 (35.1%) 14 12 Nstte e tee 19 (25.7%) 10 9tee 29 (39.2%) 18 11sPaP (mmHg) 33.6±3.5 39.8±7.7 43.6±12.9 NstaPse (mm) 19.8±2.1 17.7±3.7 17.1±3.2 NstaPse<16 21 10 11 NsLVEF<50% 14 8 6 NsrV/lV eDa area >0.6 21 9 12 NsicU Mortality 31 (41.8%) 19 12 Nstee: transesophageal echocardiography; tte: transthoracic echocardiography; sPaP: systolic pulmonary arterial pressure; taPse: tricus-pid annular Plane excursion; lVeF: left ventricular ejection fraction, rV/lV eDa: right ventricle/left ventricle end-diastolic area. group 1: normal echocardiographic findings; group 2: normal RV dimensions and function, values of sPAP≥40 mmHg; group 3: RV dilatation and values of sPAP≥40 mmHg.

Table III.—�Clinical characteristics according to echocardiographic subgroups.group 1 group 2 group 3 P

Number 25 32 17age 50.3±15.3 49.4±15.4 57.4±13.6 NssaPs ii 44.1±19.2 41.6±18.9 46.5±15.4 NsMales/females 23/2 19/13 12/5 0.02BMi 28.3±7.7 26.1±6.5 30.8±11.8 NsecMo duration days 13.5±11.1 14.5±14.2 12.1±6.6 NsMechanical ventilation days 19.9±14.4 20.0±15.9 21.5±13.2 NsicU los 23.9±17.1 23.6±19.6 23.3±13.3 NsVasopressors 14 12 11Norepinephrine only 11 9 9 NsNorepinephrine + dobutamine 3 2 1dobutamine 1 1Blood gas analysis

pH 7.38±0.11 7.35±0.10 7.36±0.11pco2 49.8±26.3 50.9±23.2 41.8±12.1 Nspo2 75.2±33.9 86.7±40.7 89.1±39.3 Nslactate (mg/dl) 2.3±1.9 3.6±3.9 3.9±4.4 Ns

ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit: SAPS II: simplified acute physiology score; BMI: Body Mass Index; LOS: length of stay. Group 1: normal echocardiographic findings; group 2: normal RV dimensions and function, values of sPAP≥40 mmHg; group 3: RV dilatation and values of sPAP≥40 mmHg.



the majority of patients (in particular in the 43.2% not associated with RV dilation and in the 23% associated with RV enlargement). So far, only few reports assessed systolic pulmo-nary arterial pressure with echocardiography in patients with acute lung injury (ali) and/or arDs 17, 21-24 and the prevalence of sPaP was not specifically addressed even if Boisser et al.17 reported that pulmonary hypertension or isolated RV dilatation was detectable in 48% of patients.

We documented for the first time that RV dilatation is an independent predictor of mor-tality in refractory arDs before VV-ecMo implantation. cepkova et al.25 assessed rV size qualitatively, “according to their laborato-ry protocol” and, among 42 patients with ali included in their investigation, rV dilation was observed in the 26% (11/42). Differently, RV dilation was quantitatively assessed and obvi-ously detected in patients with acP, indicated as rV end diastolic area alone 25 or as the ratio rV end diastolic area /lV end diastolic area.26 in arDs patients with a patent forame ovale, when compared to those without, Mekontso-Dessap 22 observed significant higher values of rVeDa/lVeDa ratio associated with higher sPaP and, similarly, legras et al. 23 reported that arDs with acP and PFo exhibited the highest rVeDa/lVeDa ratio. interestingly Boissier et al. 17 observed that moderate rV dilation (defined as RV/LD end diastolic area ratio >0.6 and <1) was detectable in the 49% of patients without acP, thus suggesting that rV dilation may precede acP development. in agreement with the study be Boissier et al.,17

as shown in table iV, the percentage of patients with taPse<16 mm was highest in group 3 who showed the lowest values of taPse. similarly the highest percentage of patients with LVEF<50% was observed in group 3 who exhibited the highest mortality rate. a patent forame ovale was detectable in 2 patients (group 2).

at stepwise regression analysis the fol-lowing variables were independent predic-tor for in-icU mortality (when adjusted for taPse<16 mm): rVeDa/lVeDa (or 0.21, 95%CI 0.062-0.709, P=0.012), BMI (OR 0.87, 95%CI 0.802-0.958, P=0.004). Hosmer-Lem-eshow χ2 7.3, P=0.50; Nagelkerke pseudo-R2 0.29.

Discussion

The main findings of the present investiga-tion, performed in 74 consecutive patients with refractory arDs before VV-ecMo implanta-tion, were as follows: 1) echocardiographic alterations were common in these patients, mainly represented by systolic pulmonary hypertension associated or not with rV dila-tation; 2) rV dilatation and BMi were inde-pendent predictors of in-icU mortality. in par-ticular BMi was inversely related to mortality.

In our series, normal findings at echocardio-graphic examination were observed in about one third of patients (33.8%), while abnor-malities were detectable in the remaining two third (66.2%). In particular, systolic arterial hypertension was the most common altera-tion at echocardiography, being detectable in

Table IV.—�Echocardiographic findings according to echo-groups.group 1 group 2 group 3 P

Echocardiographic findingstaPse (mm) 19.8±2.1 17.5±2.3 13.2±3.4* <0.001taPse < 16 0 4 (4/28, 12.5%) 17/0 (100%)* <0.001rV/lV eDa ratio 0.43±0.1 0.48±0.11 0.79±0.1* <0.001sPaP 33.6±3.5 42.5±4.9* 51.2±15.1* <0.001LVEF <50% 1 (1/24, 4%) 6 (6/26, 23.1%) 7 (7/10, 70%) * 0.02icU Mortality 11 (11/14, 44%) 9 (9/32, 28.1%) 11 (11/17, 64.7%)** 0.04tee: transesopheageal echocardiography; tte: transthoracic echocardiography; sPaP: systolic pulmonary arterial pressure; taPse: tricus-pid annular Plane excursion; lVeF: left ventricular ejection fraction, rV/lV eDa: right ventricle/left ventricle end-diastolic area. group 1: normal echocardiographic findings; group 2: normal RV dimensions and function, values of sPAP≥40 mmHg; group 3: RV dilatation and values of sPAP≥40 mmHg*= vs. group 1 P<0.001; **P<0.05 vs. group 2.



Conclusions

in conclusions, our data strongly suggest that, in patients with refractory arDs requir-ing VV-ecMo, echocardiography helps to a more accurate risk stratification, since the detection of RV dilatation identifies a subset of patients at higher risk for early death. our results, obtained from a preliminary single center study, need to be confirmed in a larger cohort of patients.

Key messages

— to date, there is a paucity of data on cardiac function assessed by echocardiog-raphy in patients with arDs unresponsive to conventional therapy submitted to VV-ecMo.

— in our series, echocardiographic al-terations were common, mainly represent-ed by systolic sPaP (associated or not with rV dilatation) and rV dilatation (together with BMi) was an independent predictor for in-icU mortality.

— our data strongly suggest that, in pa-tients with refractory arDs requiring VV-ecMo, echocardiography helps to a more accurate risk stratification.

References

1. Jardin F, Gueret P, Dubourg O, Farcot JC, Margairaz A, Bourdarias JP. Two-dimensional echocardiographic eval-uation of right ventricular size and contractility in acute respiratory failure. Crit Care Med 1985;13:952-6.

2. repessé X, charron c, Vieillard-Baron a. acute cor Pulmonale in arDs: rationale for Protecting the right Ventricle. Chest 2015;147:259-65.

3. chiumello D, Pesenti a. the monitoring of acute cor pulmonale is still necessary in “Berlin” arDs patients. Intensive Care Med 2013;39:1864-6.

4. Vieillard-Baron a. assessment of right ventricular func-tion. Curr Opin Crit Care 2009;15:254-60.

5. Khrishnan S, Schmidt GA. Acute right ventricular dys-function. a real-time management with echocardiogra-phy. Chest 2015;147:835-46.

6. combes a, ranieri M. rescue therapy for refractory arDs should be offered early: yes. intensive care Med 2015;41:923-5.

7. combes a, Brodie D, Bartlett r, Brochard l, Brower r, conrad s, et al; the international ecMo Network (ec-MoNet). Position Paper for the organization of extra-corporeal Membrane oxygenation Programs for acute

we observed that rV dilatation can be detected in about one third of patients (28.7%).

according to our data, on a clinical ground, the detection of rV dilatation at echocar-diography identifies a subset of patients with refractory arDs at higher risk of death, de-spite VV-ecMo support. Further studies are needed to identify the mechanism(s) account-ing for the development of rV dilatation and, afterwards, its management.

in our series, the incidence of acP is quite low in respect to previous papers, performed both before and in the era of protective ven-tilation.2, 5, 19, 21 it can be hypothesized that in our population acP had not enough time to develop since the prompt alert of the ecMo team and the VV- ecMo implantation due to the severe (and/or worsening) respiratory con-ditions of our patients. However, our data need to be confirmed in larger series of refractory arDs patients.

Furthermore, we documented that BMi is an independent predictor of in icU death in refractory arDs treated with VV-ecMo and that, in particular, lower BMi was associated with higher mortality. our results are in keep-ing with those reported by Al-Soufi S et al. 28 who, in a retrospective analysis of the interna-tional extracorporeal life support organiza-tion enrolling 1334 adult patients supported by VV-ecMo, documented that increased BMi was not a risk factor for in-hospital mortality (even if at univariate analysis increased body weight was associated with a reduced risk of death).

Limitations of the study

a limitation of the present study may be rep-resented by the fact that we report only on one echocardiographic examination (just before ecMo implantation). Data on serial echo-cardiographic exams during ecMo support might provide information potentially use-ful for intensivists. in our series, lV ejection fraction was calculated by eyeball, a method which can be performed more rapidly and has been shown a good correlation with the simp-son method in critically ill patients.27



18. Jardin F, Dubourg O, Bourdarias JP. Echocardiographic pattern of acute cor pulmonale. Chest 1997;111:209-17.

19. Lhéritier G, Legras A, Caille A, Lherm T, Mathonnet A, Frat JP, et al. Prevalence and prognostic value of acute cor pulmonale and patent foramen ovale in ventilated patients with early acute respiratory distress syndrome: a multicenter study. Intensive Care Med 2013;39:1734-42.

20. Muller r, Buttner P. a critical discussion of intra-class correlation coefficients. Statistics in Medicine 1994;13:2465-76.

21. Vieillard-Baron A, Schmitt JM, Augarde R, Fellahi JL, Prin s, Page B, et al. acute cor pulmonale in acute res-piratory distress syndrome submitted to protective ven-tilation: incidence, clinical implications, and prognosis. Crit Care Med 2001;29:1551-5.

22. Mekontso Dessap a, Boissier F, leon r, carreira s, campo Fr, lemaire F, et al. Prevalence and prognosis of shunting across patent foramen ovale during acute respir-atory distress syndrome. Crit Care Med 2010;38:1786-92.

23. Legras A, Caille A, Lheritier G, Frat J, Mathonnet A, courte-rabiller a, et al. transthoracic and transesopha-geal echocardiography during acute respiratory distress syndrome: incidence of acute cor pulmonale and patent foramen ovale. arcoFoP multicenter study-prelimi-nary results. Arch Cardiovasc Dis 2011;104:274-75.

24. Lai PS, Mita C, Thompson BT. What is the clinical sig-nificance of pulmonary hypertension in acute respira-tory distress syndrome? a review. Minerva anestesiol 2014;80:574-85.

25. Cepkova M, Kapur V, Ren X, Quinn T, Zhou H, Foster e, et al. Pulmonary dead space fraction and pulmonary artery systolic pressure as early predictors of clinical out-come in acute lung injury. chest 2007;132:836-42.

26. Vieillard-Baron a, charron c, caille V, Belliard g, Page B, Jardin F. Prone positioning unloads the right ventricle in severe arDs. chest 2007;132:1440-6.

27. Al-Soufi S, Buscher H, Nguyen ND, Rycus P, Nair P. lack of association between body weight and mortal-ity in patients on veno-venous extracorporeal membrane oxygenation. Intensive Care Med 2013;39:1995-2002

28. Bergenzaun L, Gudmundsson P, Öhlin H, Düring J, Ers-son a, ihrman l, et al. assessing left ventricular systolic function in shock: evaluation of echocardiographic pa-rameters in intensive care. Crit Care 2011;15:R200.

Respiratory Failure in Adult Patients. Am J Respir Crit Care Med 2014;190:488-96.

8. schmidt M, Hodgson c, combes a. extracorporeal gas exchange for acute respiratory failure in adult patients: a systematic review. Critical Care 2015;19:99.

9. Peris A, Lazzeri C, Cianchi G, Bonizzoli M, Batacchi S, Bernardo P, et al. Clinical Significance of echocardiogra-phy in patients supported by venous-venous extracorpor-eal membrane oxygenation. J Artif Organs 2015;18:99-105.

10. ARDS Definition Task Force, Ranieri VM, Rubenfeld gD, et al. acute respiratory distress syndrome: the Berlin Definition. JAMA 2012;307:2526-33.

11. cianchi g, Bonizzoli M, Pasquini a, Bonacchi M, Zagli g, ciapetti M, et al. Ventilatory and ecMo treatment of H1N1-induced severe respiratory failure: results of an italian referral ecMo center. BMc Pulm Med 2011;11:2.

12. ciapetti M, cianchi g, Zagli g, greco c, Pasquini a, spina r, et al. Feasibility of inter-hospital transportation using extra-corporeal membrane oxygenation (ecMo) support of patients affected by severe swine-flu(H1N1)-related ARDS. Scand J Trauma Resusc Emerg Med 2011;19:32.

13. Le Gall JR, Lemeshow S, Saulnier F. A new simpli-fied acute physiology score (SAPS II) based on a Eu-ropean / North American multicenter study. JAMA 1993;270:2957-63.

14. lazzeri c, gensini gF, Peris a.the assessment of cardiac function in veno-venous extracorporeal membrane oxy-genation: the emerging role of bedside echocardiography. Heart Lung Vessel 2015;7:99-100.

15. Lang R, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, ernarnde l et al. recommendations for cardiac cham-ber Quantification by Echocardiography in Adults: An Update from the american society of echocardiography and the european association of cardiovascular imag-ing. J Am Soc Echocardiogr 2015;28:1-39.

16. Fagnoul D, Pasquier P, Bodson L, Ortiz JA, Vincent JL, De Backer D. Myocardial dysfunction during H1N1 in-fluenza infection. J Crit Care 2013;28:321-7.

17. Boissier F, Katsahian S, Razazi K, Thille AW, Roche-campo F, leon r, et al. Prevalence and prognosis of cor pulmonale during protective ventilation for acute respiratory distress syndrome. intensive care Med 2013;39:1725-33.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: March 8, 2016. - Manuscript accepted: March 4, 2016. - Manuscript revised: February 3, 2016. - Manu-script received: September 16, 2015.


Usa), a vl with an angulated blade, tended to be one of the most extensively studied vl for place-ment the Dlt.4, 6, 7 However, the extreme angula-tion of these designs may complicate advancing the Dlt towards the glottis opening, despite they may provide superior laryngeal visualization.

the channeled yellow airtraq® (Prodol Med-itec s.a., vizcaya, spain) and standard non-channeled blade of the King vision™ (ambu, Ballerup, copenhagen, Denmark) may offer ad-ditional benefits for DLT intubation in patients

several regional surveys demonstrated that most thoracic anesthesiologists are using

the double-lumen tubes (DLT) as the first-choice lung separation technique.1-3 Dlt, when compared with single lumen tracheal tube, can be more difficult to insert in patients with dif-ficult airways. The videolaryngoscopes (VL) have the potential to facilitate the placement of the Dlts for lung separation in patients with potential difficult airway.4, 5

the glidescope® (verathon inc., Bothell, Wa,


comparison of three videolaryngoscopes for double-lumen tubes intubation in simulated

easy and difficult airways: a randomized trialMohamed r. el-taHan *, abdulmohsen a. al’gHaMDi,

alaa M. KHiDr, ihab s. gaaroUr

Department of anesthesiology, King Fahd Hospital, University of Dammam, Dammam, saudi arabia*Corresponding author: Mohamed R. El-Tahan, Anesthesiology Department, King Fahd Hospital of the University of Dammam, Dammam, Saudi Arabia, P.O. 40289 Al Khubar 31952, Saudi Arabia. E-mail: [email protected]


Lavoro: 11343-MAStitolo breve: VIDEOLARYNGOSCOPE COMPARISON FOR DOUBLE-LUMEN TUBES intUBationprimo autore: EL-TAHANpagine: 1050-8citazione: Minerva Anestesiol 2016;82:1050-8

a B s t r a c tBACKGROUND: The King Vision™ (KVL) and Airtraq® videolaryngoscopes may reduce the time to double lumen tube (Dlt) intubation compared to the glidescope® and MacIntosh in simulated easy and difficult airways.METHODS: Twenty-one staff anesthesiologists with limited prior experience in using videolaryngoscopes for DLT in-tubation were assigned randomly to insert a Dlt using the Macintosh, glidescope®, airtraq® and Kvl videolaryngo-scopes on easy and difficult airway simulators in a randomized crossover order. Time to DLT intubation, laryngoscopic view, intubation difficulty, optimizing manoeuvers and failure to intubation — defined as an attempt taking longer than 150 s — were recorded.RESULTS: The three videolaryngoscopes had comparable times to intubation and glottis visualization in both scenarios. compared with the Macintosh, the Kvl had longer intubation times in the simulated easy airway scenario (mean 9.2 vs. 21.1 s, respectively, P<0.001). in both scenarios, the airtraq® took a longer intubation time than the Macintosh (P<0.001 and P=0.019, respectively). the glidescope® was easier to use than the airtraq® and Kvl in the easy airway scenario (P=0.021 and P=0.001, respectively). The KVL had higher intubation difficulty scores than the GlideScope® and airtraq® (P=0.002 and P=0.008, respectively) in both scenarios and required more frequent optimizing manoeuvers than the gli-descope® (P=0.012) in the simulated easy airway. Two participants failed to intubate the difficult airway simulator; one with the Macintosh and the other with the Kvl.CONCLUSIONS: The Airtraq® and non-channeled Kvl required more time over the Macintosh for Dlt intubation, as a primary outcome, but the success rates for the 3 videolaryngoscopes were very high.(Cite this article as: El-Tahan MR, Al’Ghamdi AA, Khidr AM, Gaarour IS. comparison of three videolaryngoscopes for dou-ble-lumen tubes intubation in simulated easy and difficult airways: a randomized trial. Minerva Anestesiol 2016;82:1050-8)Key words: laryngoscopes - video-assisted surgery - intratracheal intubation.


VIDEOLARYNGOSCOPE COMPARISON FOR DOUBLE-LUMEN TUBES INTUBATION el-taHan


We hypothesized that the use of the airtraq® and the standard non-channeled blade of King vision™ would be associated with shorter times to Dlt intubation compared with the Macintosh and glidescope® laryngoscopes. We have considered to assess the efficacy of each device on manikins before considering to evaluate them in patients undergoing thoracic procedures.


according to the research committee, an-esthesiology Department King Fahd Hospital of the University of Dammam, saudi arabia on 20th november 2014, we did not require approval from an institutional ethical review committee. this study was retrospectively reg-istered with www.clinicaltrials.gov (Identifier nct02640196). Following written consent, 21 staff anesthesiologists at the authors’ department

with limited mouth opening or restricted neck movement,8-10 in whom the use of traditional vl like as the glidescope® or Mcgrath® could be difficult. This is because they have less an-gulated blades and offer an adjacent passage to guide the large outer diameter and more rigid design of Dlts.11 compared with the chan-nelled airtraq® and the airway scope®, the standard non-channelled blade King vision™ has the thinnest and shortest stature (26 vs. 28 mm, and 49 and 13 mm vs. 18 and 131 mm, respectively) and the widest field of view (160° vs. 80° and 90°, respectively).10 thus, it is like-ly that the use of standard non-channeled blade King vision™ may be associated with better Dlt intubation conditions.

to the best of our knowledge, the compari-son of the effects of the Macintosh, glides-cope®, airtraq®, and King vision™ vl on the time to Dlt intubation in simulated easy and difficult airways has not yet been studied.

Figure 1.—A) The MacIntosh, reusable GlideScope® gvl, the yellow airtraq® and standard non-channeled King vision™ blades; B) the stylet of a 35-Fr left-side double-lumen tube is bent to fit the natural curve of the MacIntosh, reusable Gli-descope® GVL, and standard non-channeled King Vision™ blades; whereas the DLT is loaded through the lumen of the airtraq® after removal of the stylet.

a

B

el-taHan VIDEOLARYNGOSCOPE COMPARISON FOR DOUBLE-LUMEN TUBES INTUBATION


Head rest, adult (140 mm in height) (covi-dien, Mansfield, MA, USA) under the occiput and securing the head position with adhesive tape, object to replicate cervical-collar use (Figure 2B). Positioning was confirmed after each attempt to ensure consistency.

Before the study, all participants received a standardized demonstration on the study pro-tocol regarding the correct use of each device for placement of Dlt on the “easy” airway simulator, conducted by M-re (>5 years of experience in using the vls for Dlt intuba-tion). then each participant was then allowed two practice attempts at Dlt intubation with each device, one in a simulated “easy” airway and second in a simulated “difficult” airway, with the supervision of M-re who gave them a constructive feedback.

Before each Dlt intubation attempt, the manikin, laryngoscope blade and Dlt were lubricated. Participants were randomly al-located by M-re into four groups according to the first intubation method to be performed and the order in which to apply the different laryngoscopes within each group into a simu-lated “easy” airway by drawing sequentially numbered sealed opaque envelopes contain-ing a computer-generated randomization code groups based on circular permutation (Figure 3). The first group attempted DLT intuba-tion using the Macintosh (blade size4) laryn-goscope, the second using the glidescope®

participated in this randomized, controlled, crossover manikin study. they were informed that they could withdraw from the study at any time and the data on the individual performance were not made available to anyone outside the study team. all anesthesiologists involved in this study were familiar with Dlt insertion on a regular basis. all participants had previous ex-perience with the 3 tested vls (Figure 1) for tracheal intubation; however, the vast majority of them had only used the reusable standard glidescope® gvl and the single-use yellow airtraq® and standard non-channeled King vi-sion™ vls (Figure 1a, B) for Dlt insertion around one to 10 times (table i).

Two high-fidelity simulators (Airway Man-agement trainer, model aa-3100, laerdal Medical ltd., orpington, england, UK) were prepared to simulate easy and difficult airway situations, as described by Wang et al.13 and Marshall et al.14 the “easy” airway was estab-lished with the manikin in a neutral position (Figure 2A). The “difficult” airway setting was obtained by placing an oasis elite™ Prone

Figure 2.—A simulated “easy-airway” manikin (A) and “difficult-airway” manikin (B) which was obtained by placing an adult Prone Head rest under the occiput (arrow 1) and securing the head position with adhesive tape to replicate cervical-collar use (arrow 2).

Table I.—�Previous experience of the 21 participants with using each device.

Macintosh glidescope® airtraq® King vision™

< 10 uses 0 19 * 19 * 20 *≥ 10 uses 21 2 * 2 * 1 *Data are presented as numbers.*P<0.0002 significant compared with the MacIntosh laryngoscope.

a B



blade or Dlt or to ask the supervising investi-gator (aMK, isg) to help solve the problem, and gave suggestions or instructions. an in-vestigator (aMK, isg) was available to hand them the Dlt, remove the stylet after place-ment, and inflate the tube cuff.

With each device in each scenario, the pri-mary endpoint was the time to achieve suc-cessful DLT intubation, defined as the time when the investigated laryngoscope passed the central incisors to when the tip of the bronchial lumen passed through the glottis, as confirmed visually by the operator (in the Macintosh group) or by the investigator, thanks to display screens (in the vls groups).12 the success of Dlt intubation in the Macintosh group, was accepted upon declaration of one of the inves-tigators (aMK, isg) with the use of the glide-scope®. secondary outcomes included the best view during laryngoscopy using the classifica-tion described by Cormack and Lehane;16 the difficulty of intubation, first-pass success cal-culated as number of first-attempt successes/number of intubation attempts, and number of times that the optimization manoeuvers were required. additionally, at the end of their par-

gvl (blade size 4), the third using the yellow airtraq®, and the fourth using the standard non-channeled King vision™ (size 3). the second and the following devices were determined in the same order in every group as shown in Fig-ure 3. next, in the same random order, they used the 4 devices into the simulated “diffi-cult” airway conditions. after completing the Dlt intubation, participants had a 15-minute break before performing intubation using an-other laryngoscope. all intubations were per-formed with a 35-Fr left Dlt. the participants were not allowed to watch each other to avoid any learning effect through observation.

in the glidescope® group, the Dlt was in-serted as described by Hsu et al.15 the airtraq® aWDr video system was used to make sure that the view of the glottis could be shared be-tween operator and investigator. We described earlier four necessary steps to insert a Dlt us-ing a non-channelled blade of King vision™.10

When the anesthesiologists encountered dif-ficulty in visualising vocal cords or placing the Dlt, they were allowed to use any manoeuver they would normally use to navigate the Dlt into the trachea including readjustment of the

Figure 3.—Study flow diagram.

(N.=6)(N.=6)

(N.=0)

(N.=5)(N.=5)

(N.=0)

(N.=5)(N.=5)

(N.=0)

(N.=5)(N.=5)

(N.=0)

(N.=6)(N.=0)

(N.=5)(N.=0)

(N.=5)(N.=0)

(N.=5)(N.=0)

(N.=6)(N.=0)

(N.=5)(N.=0)

(N.=5)(N.=6)

(N.=5)(N.=0)

(N.=6)(N.=0)

(N.=5)(N.=0)

(N.=5)(N.=0)

(N.=5)(N.=0)

(N.=21)(N.=0)

(n.=21)

(n.=2)(n.=2)

(n.=23)



ber (%). a P value of less than 0.05 was consid-ered as statistically significant.

Results

twenty three participants (23 male, median age 36 [27-59] years) were assessed for eligi-bility; two declined to participate. All remain-ing 21 participants intubated the two simulated settings with each device (Figure 3). Partici-pants’ prior experience in using the three vls in Dlt intubations are presented in table i.

the three vl had comparable times to Dlt intubation (table ii, Figure 4), views obtained at laryngoscopy and success rates in both simulated easy and difficult airway scenarios (table ii).

the airtraq® took a significantly longer time to Dlt intubation than the Macintosh in both scenarios (mean time; easy settings 21.3 s [95% CI: 16.51 to 26.10] vs. 9.2 s [95% CI: 7.20 to 11.32], P<0.001; difficult settings 25.7 s [95% CI: 19.51 to 31.95] vs. 16.8 s [95% CI: 13.80 to 19.73], P=0.019) (table ii, Figure 4). compared with the Macintosh, the King vi-sion™ had significantly longer times to DLT intubation in the simulated easy airway sce-nario (mean 21.1 s [95% CI: 16.91 to 25.41] vs. 9.2 s [95% CI: 7.20 to 11.32]) (P<0.001) (table ii, Figure 4).

the participants found the glidescope® la-ryngoscope significantly easier to use than the airtraq® and King vision™ vl in the easy air-way scenario (P=0.021 and P=0.001, respec-tively) (table2). the use of King vision™ re-sulted in a higher score of difficult intubation than the glidescope® and airtraq® in the sim-ulated difficult airway scenarios (P=0.002 and P=0.008, respectively). there was a statistical significant difference in number of optimiza-tion manoeuvers between the King vision™ and glidescope® in the easy airway scenario (P=0.012) (table ii).

two participants failed to intubate the simulated difficult airway manikin; one with the Macintosh and the other with the King vision™. Failure was due to the inability to navigate the Dlt through the glottis opening. When participants were asked which laryngo-

ticipation in the trial, the participants stated which device they preferred after having used the 4 of them.

The difficulty of intubation was evaluated using a visual analog scale (vas) (ranging from 0, meaning extremely easy, to 100, ex-tremely difficult) expressed by the anesthesi-ologists after the Dlt intubation. additionally, the failure rate for DLT intubation, defined as an attempt taking longer than 150 seconds, was recorded.15 the cause of failure was also documented.

the previous Dlt intubation experience with the vls was recorded. all data, with the exception of the DLT intubation difficulty score, were determined by an independent in-vestigator. the participants were not informed about the time taken to achieve any intuba-tions.


a pilot study showed that the mean time to Dlt intubation using the glidescope® was 57.8s with a standard deviation of 4.10 s. an a priori power analysis indicated that a sample size of 16 participants was sufficiently large to detect a 10-second difference in the time to Dlt intubation, which was assumed to be of clinical importance during the use of the chan-neled VL, a type I error of 0.008 (0.05/6 pos-sible comparisons) and a power of 90%. We added more operators (20%) for a final sample size of 21 participants to compensate dropping out during the study.

statistical comparisons were restricted to be-tween laryngoscope analyses within each sce-nario. Data were tested for normality using Kol-mogorov-smirnov test. Data for the duration of DLT intubation and the intubation difficulty score were analyzed using one way analysis of variance (anova) on ranks with post-hoc stu-dent-newman-Keuls tests. the Kruskal-Wallis one-way anova was performed to compare the non-parametric values of the 4 groups and post hoc pairwise comparisons were performed using the Wilcoxon rank sum t-test. categori-cal variables were compared using Mcnemar’s test. Data are presented as mean (sD) or num-



success rates. this illustrates how the use of the vls can be useful for Dlt intubation particu-larly in the difficult airway settings. In contrast to others,11, 17 who have reported a shorter or a comparable time to Dlt intubation in humans, in the present manikin study, the airtraq® took a longer time to Dlt intubation than the Macin-tosh in both scenarios. this may be due to the diversity of prior operator experience compared with these studies. However, the differences of 9s to 12s between the two devices may be not

scope they would prefer in real-life Dlt intu-bation, 100% detected the glidescope®.

Discussion

the authors showed that, in simulated easy and difficult airway settings, the use of the glidescope®, airtraq®, and King vision™ for insertion of Dlt by less experienced anesthe-siologists was associated with comparable intu-bation times, glottis visualization, and first pass

Figure 4.—Time to double-lumen tube (DLT) intubation in the simulated easy- and difficult-airway settings. Data are pre-sented as mean (sD). *P<0.05 vs. the Macintosh group.

Table II.—�Outcome data.Macintosh

(n.=21)glidescope®

(n.=21)airtraq®

(n.=21)King visiontM

(n.=21) P value

easy-airway scenariotime to Dlt intubation (s) 9.2 (4.58) 15.2 (5.31) 21.3 (10.56) * 21.1 (9.31) * <0.001Cormack Lehane score (I/II/III/IV) 21/0/0/0 21/0/0/0 21/0/0/0 21/0/0/0Score of difficult intubation 26 (16.3) 20 (8.10) 30 (13.0) † 35 (14.4) † 0.002First-pass success (%) 100% 100% 100% 100%Number of laryngoscopy attempts (1/2/3) 21/0/0 21/0/0 21/0/0 21/0/0number of optimization manoeuvers 2 [2-4] 2 [2-3] 3 [2-4] 4 [2-4] † 0.012

Difficult-airway scenariotime to Dlt intubation (s) 16.8 (6.53) 19.9 (7.70) 25.7 (13.71) * 21.5 (8.71) 0.027Cormack Lehane score (I/II/III/IV) 0/16/5/0 0/18/3/0 3/16/2/0 3/16/2/0 0.106Score of difficult intubation 35 (19.7) 23 (13.5) 25 (13.3) 37 (9.6) † ‡ 0.002First-pass success (%) 95.4% 100% 100% 95.4% 0.999Number of laryngoscopy attempts (1/2/3) 20/0/1 21/0/0 21/0/0 20/1/0 0.559number of optimization manoeuvers 3 [2-4] 3 [2-4] 2 [2-3] 3 [2-3] 0.667

Data are presented as mean (sD), number, proportion, or median [interquartile range].*P<0.05 vs. the Macintosh, † vs. the glidescope®, or ‡ vs. the airtraq®.



Dlt intubation. thus, the three vls devices have advantages when Dlt intubation is not straightforward. However, an improved view of the glottis does not always translate into eas-ier Dlt intubation. similarly, previous reports observed comparable glottic visualization dur-ing the use of Macintosh and airtraq®.11, 19 the use of King vision™ required more frequent optimizing manoeuvers than the glidescope®, despite comparable visualization of glottis. this could be explained by the occasional problems related to the Dlt delivery and ad-vancement into the trachea alongside the non-channeled King vision™ blade. By contrast, the channeled airtraq® has a channel to guide the DLT; thus, once an adequate view of the glottis has been obtained, the vl is kept steady and the Dlt advanced into the glottis.

in similar to others who have reported overall successful intubation rates of 97% to 100% for the glidescope®,6, 18 and 94% for the airtraq®,18 in the present study all intuba-tions were successful in the glidescope® and airtraq® groups. However, it is much higher than the 83%success rate reported by russell et al.7 (P<0.001 as tested using χ2 test). in that study, the failure rate may be partially inflated by the tight definition of DLT intubation fail-ure, including a 120-s intubation time limit.7 these studies 6, 7, 18 as well as our study, how-ever, were not powered to test this difference. in our small sample, the failure rate of 4.8% at the first attempt using the MacIntosh and King vision™ was due to the impaction of the bron-chial lumen of the Dlt on the vocal cords or anterior larynx. in the two failures, we would not identify if the anesthesiologists able to in-tubate with another tool after the failure.

our failure to reject the null hypothesis could be explained by the inherent problems in all manikin studies in that the times required to perform intubation are generally quicker than in patients. although the present results has a limited clinical importance, our results may therefore have greater applicability to the ma-jority of anesthesiologists who are still learn-ing to use the three devices for Dlt intubation, and we have shown that the acute-angle gli-descope® was easiest to use in both simulated

clinically relevant. the King vision™ was as-sociated with a longer time to Dlt intubation than the Macintosh because the non-channeled design of the King vision™ has potentially increased the difficulty in passage of the DLT through the trachea. Unfortunately, our result cannot be extrapolated to other vl devices like the c-Mac, Mcgrath® and airway scope®. However, this study is not meant to be a com-prehensive review of the performance of all vls, but is representative of those commonly available at the authors’ center.

in our study, the glidescope® was the easi-est to use in both scenarios. this may be ex-plained by the familiarity of the participants with using the angulated blade design for tra-cheal intubation.7 additionally, the more rigid design of Dlts make them relatively harder to insert it through the channeled airtraq® blades or alongside the non-channeled King vision™ blade.10 in contrast, other investigators 7 found the glidescope® more difficult to use than the Macintosh laryngoscope for Dlt intubation. that study, however, was done on patients with no predictors for difficult laryngoscopy. Other investigators reported comparable diffi-cult intubation scores with the use of the gli-descope® and airtraq®, when used by expe-rienced laryngoscopists with over 30 times,18 whereas the vast majority of operators in the present study has limited experience with us-ing the airtraq® (<10 times).

the glidescope®, yellow airtraq® and the non-channelled King vision™ accommodate a minimum mouth opening of 25mm, 18mm and 13mm, respectively. We demonstrated before the successful use of the standard non-channelled blade of the King vision™ for placement of a Dlt in a morbidly obese pa-tient with predicted difficult airway.10in the present study, the difficulty encountered with the King vision™ may be related to the fact that the operator is closer to the screen and loses the comfort of a more distant and bigger screen as with the glidescope® and airtraq®.

the authors reported that all 4 devices dis-play the larynx (12/84, 14%) viewings were cormack and lehane iii and none was grade IV in the simulated difficult conditions for



more difficult to navigate through the vocal cords particularly with the yellow airtraq®.20 third, the bias of choosing the operators among less experienced anesthesiologists must certainly be taken into consideration, such a choice reduced the general validity of the re-sults and does not help the reader to understand the real value of the used devices. this makes our conclusion difficult to extrapolate to the usual thoracic anesthesiologists who nowa-days perform vl on a regular basis. However, we believe that our participants may reflect the more common clinical practice where the an-esthesiologist has more experience using the old technique, in this study, the Macintosh and glidescope® laryngoscopes. Fourth, in similar to previous reports 12, 14 we used the subjective VAS to assess the intubation difficulty rather than the validated intubation difficulty scales as the adnet’s scale or likert scale.11, 13

Conclusions

in conclusion, the authors found that, the airtraq® and non-channeled King vision™ vls required more time over the Macintosh for Dlt intubation, but the success rates for the three vls were very high.

Key messages

— the 3 videolaryngoscopes had com-parable times to Dlt intubation and glottis visualization in both scenarios.

— the airtraq® and the non-channeled blade of the King vision™ took longer times to Dlt intubation than the Macin-tosh in routine airway management.

— the glidescope® was easier to use than the airtraq® and King vision™ in the simulated routine airway.

— The KVL was more difficult to use than the other two videolaryngoscopes in both simulated scenarios

— the limited experience of the opera-tors in using the videolaryngoscopes for Dlt intubation may reduce the general va-lidity of our results.

easy and difficult intubation conditions. Addi-tionally, although we were unable to detect sta-tistical differences, the success rates for both glidescope® and airtraq® devices were very high in the simulated difficult airway condi-tions. of note, the learning curve may apply to the use of different vls as it has been shown for the use of the glidescope® alone.6, 7 Fur-ther studies in humans are required to extend these findings to the clinical settings in a mixed operators and diverse patient population. the authors assumed that a 15-minute washout pe-riod would eliminate the potential problem of carryover effects in our crossover design.

similar to previous studies 6, 7, 12, 15, 18 the primary outcome in this study was the time to DLT intubation, which reflects the period when patients are at risk of hypoxemia. the difference in intubation duration between the studied devices may not be clinically signifi-cant for normal healthy patients, but few sec-onds may worth for patients undergoing tho-racic surgery with potential risk for hypoxemia who were not studied in this study.7

the costs per unit blade of reusable glide-scope® and the single-use yellow airtraq® and standard non-channeled King vision™ were calculated at the authors’ center to be equiva-lent to Us$ 13,500, Us$ 93, and Us$ 100, re-spectively. compared with the more expensive bigger screen glidescope®, the little screen single-use airtraq® and non-channeled Kvl potentially eliminate the risk of cross-con-tamination and have comparable times to in-tubation, glottis visualization, and success rate at an affordable lower price. However, these costs may vary from one country to another.


there are several limitations to this study. First, the simulator used in this study was not a specifically designed difficult airway simula-tor and the used cervical-spine immobilization position may be not very difficult. However, the “difficult” airway setting was simulated as described earlier,13, 14 where the spine was immobilized. second, only size 35 Dlts were studied; however, larger tubes would be much



patient through the non-channelled blade of the King vi-sion(™) videolaryngoscope. F1000Res 2014;3:129.

11. Wasem s, lazarus M, Hain J, Festl J, Kranke P, roewer n, et al. comparison of the airtraq and the Macintosh laryngoscope for double-lumen tube intubation: a rand-omized clinical trial. Eur J Anaesthesiol 2013;30:180-6.

12. gamez r, slinger P. a simulator study of tube ex-change with three Different Designs of Double-lumen Tubes. Anesth Analg 2014;119:449-53.

13. Wang PK, Huang CC, Lee Y, Chen TY, Lai HY. Com-parison of 3 video laryngoscopes with the Macintosh in a manikin with easy and difficult simulated airways. Am J Emerg Med 2013;31:330-8.

14. Marshall eM, o’loughlin e, swann aD. First compari-son of the venner(™) a.P. advance(™) versus the Mac-Intosh laryngoscope for intubations by non-anaesthetists: a manikin study. Emerg Med Australas 2014;26:262-7.

15. Hsu HT, Chou SH, Chou CY, Tseng KY, Kuo YW, Chen Mc, et al. A modified technique to improve the outcome of intubation with a left-sided double-lumen endobron-chial tube. BMC Anesthesiol 2014;14:72.

16. Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia 1984;39:1105-11.

17. Hamp t, stumpner t, grubhofer g, ruetzler K, thell r, Hager H. Haemodynamic response at double lumen bronchial tube placement-airtraq vs. Macintosh laryngo-scope, a randomised controlled trial. Heart lung vessel 2015;7:54-63.

18. Yi J, Gong Y, Quan X, Huang Y. Comparison of the Air-traq laryngoscope and the glidescope for double-lumen tube intubation in patients with predicted normal air-ways: a prospective randomized trial. BMC Anesthesiol 2015;15:58.

19. Hirabayashi Y, Seo N. The Airtraq laryngoscope for placement of double-lumen endobronchial tube. can J Anaesth 2007;54:955-7.

20. el-tahan Mr. videolaryngoscopes for placement of dou-ble lumen tubes: Is it time to say goodbye to direct view? Saudi J Anesth 2016;10:29-38.

References

1. Della rocca g, langiano n, Baroselli a, granzotti s, Pravisani c. survey of thoracic anesthetic practice in Italy. J Cardiothorac Vasc Anesth 2013;27:1321-9.

2. eldawlatly a, turkistani a, shelley B, el-tahan M, Mac-fie A, Kinsella J; Thoracic-anaesthesia Group Collabora-tors. Anesthesia for thoracic surgery: a survey of Middle Eastern practice. Saudi J Anaesth 2012;6:192-6.

3. Shelley B, Macfie A, Kinsella J. Anesthesia for thoracic surgery: a survey of UK practice. J Cardiothorac Vasc Anesth 2011;25:1014-7.

4. Chen A, Lai HY, Lin PC, Chen TY, Shyr MH. Glide-scope-assisted double-lumen endobronchial tube place-ment in a patient with an unanticipated difficult airway. J Cardiothorac Vasc Anesth 2008; 22:170-2.

5. campos JH. lung isolation techniques for patients with difficult airway. Curr Opin Anaesthesiol 2010;23:12-7.

6. Hsu HT, Chou SH, Wu PJ, Tseng KY, Kuo YW, Chou CY, et al. comparison of the glidescope® videolaryngoscope and the Macintosh laryngoscope for double-lumen tube intubation. Anaesthesia 2012;67:411-5.

7. russell t, slinger P, roscoe a, Mcrae K, van rens-burg a. a randomised controlled trial comparing the glidescope((r)) and the Macintosh laryngoscope for double-lumen endobronchial intubation. anaesthesia 2013;68:1253-8.

8. liu eH, goy rW, tan BH, asai t. tracheal intubation with videolaryngoscopes in patients with cervical spine immobilization: a randomized trial of the Airway Scope and the GlideScope. Br J Anaesth 2009;103:446-51.

9. salazar Herbozo e, Planas B, ramasco F, gómez rice a, catalán P. Double lumen tube insertion in awake pa-tients through the airtraq laryngoscope in 2 cases of expected difficult airway. Rev Esp Anestesiol Reanim 2011;58:315-7.

10. el-tahan M, Doyle DJ, Khidr aM, Hassieb ag. case Report: Double lumen tube insertion in a morbidly obese

Congresses.—the material discussed in this paper has been presented as an oral presentation at the 31st annual congress of the euro-pean association of cardiothoracic anesthesiologists, which was held in Basel, switzerland, on May 11th-13th, 2016.Conflicts of interest.—Abdulmohsen A. Al’Ghamdi, Alaa M. Khidr and Ihab S. Gaarour declare that they have no conflicts of interest and that they have received no financial support for the research, authorship, and/or publication of this article. Mohamed R. El Tahan received free airway device samples from ambu in april 2014 for use in another study, and from airtraq in March 2015 for use in the present study and he has no direct financial or other interest in Ambu or Airtraq (in the context of this and other studies).Acknowledgements.—the authors wish to express their appreciation to the staff anesthesiologists at the Department of anesthesiol-ogy, King Fahd Hospital of the University of Dammam, for their involvement and participation in this study. also, the authors thank Professor Peter slinger, from the toronto general Hospital, for reviewing the protocol and preliminary draft of the manuscript.Article first published online: May 12, 2016. - Manuscript accepted: May 11, 2016. - Manuscript revised: May 6, 2016. - Manuscript received: March 21, 2016.


the opposite effect by enlarging the interstitial space thus reducing oxygen diffusion in tissues involved in systemic inflammation. To predict the cardiac output response to fluid loading and avoid a fluid overload, particularly in pa-tients in whom a preload increase leaves co unchanged, many investigators have proposed passive leg raising (Plr) 1-4 as a manoeuver to be used in ventilated or spontaneously breath-ing patients.

critically ill patients customarily receive fluid therapy aimed to increase cardiac

output (co) and guarantee blood perfusion to organs and tissues. Even though fluid infu-sion could increase co in responder patients no conclusive evidence exists that this he-modynamic change induces a corresponding increase in overall organ perfusion or causes


effects of passive leg raising on microvascular venous compartment in critically ill patients

Roberto A. DE BLASI *, Roberto ARCIONI, Domitilla BRANCADORO, Monica ROCCO

Intensive Care Unit, Department of Medical and Surgical Science and Translational Medicine, Faculty of Medicine and Psychology, University of Rome “Sapienza”, Rome, Italy*Corresponding author: Roberto A. De Blasi, Intensive Care Unit, Department of Medical and Surgical Science and Translational Medicine, Faculty of Medicine and Psychology, University of Rome “Sapienza”, Rome, Italy. E-mail: [email protected]


lavoro: 10922-Mastitolo breve: MICROVASCULAR BED CHANGES AFTER PASSIVE LEG RAISINGprimo autore: De Blasipagine: 1059-68citazione: Minerva Anestesiol 2016;82:1059-68

a B s t r a c tBACKGROUND: Even though fluid loading is thought to improve organ perfusion, the way in which it does so remains unclear. We assessed how the microvascular bed in skeletal muscle reacts to passive leg raising in patients with and without sepsis or septic shock.MetHoDs: We studied 40 critically ill patients (group a) and 30 healthy controls (group B). the forearm microvas-cular bed was assessed using near-infrared spectroscopy before and after passive leg raising. We measured stressed and unstressed volumes, inside pressures, blood flow, microvascular compliance and tone.RESULTS: In group A, passive leg raising induced a microvascular bed increase from 4.9 (3.2-6.5) mL/100 mL tissue to 5.7 (3.9-8.1) mL/100 mL tissue (P=0.005), leaving inside pressures unchanged, whereas in group B neither volumes nor pressures changed. Patients without sepsis showed an increase in the stressed volume from 0.22 (0.10-0.28) mL/100 mL tissue to 0.34 (0.23-0.66) mL/100 mL tissue (P=0.039) and a decrease in compliance (P=0.004), whereas, in septic shock, the unstressed volume increased from 4.20 (3.01-5.82) mL/100 mL tissue to 5.32 (4.01-11.50) mL/100 mL tissue (P=0.036). In critically ill patients near-infrared spectroscopy showed no difference in microvascular variables between responders and non-responders to passive leg raising, but responders showed a cardiovascular response shorter than healthy subjects.coNclUsioNs: our study provides evidence that macrocirculatory parameters are unreliable to derive measurements of stressed and unstressed volumes. Our results indicate that in septic shock, the enlargement of the unstressed volume associated with passive leg raising induces loss of fluids to the interstitium, thus leaving organ perfusion unchanged or worse.(Cite this article as: De Blasi RA, Arcioni R, Brancadoro D, Rocco M. Effects of passive leg raising on microvascular venous compartment in critically ill patients. Minerva Anestesiol 2016;82:1059-68)Key words: skeletal muscle - Venules - Blood volume determination - Venous pressure - Near-infrared spectroscopy.

comment in p. 1026.

Minerva Anestesiologica 2016 October;82(10):1059-68© 2016 eDiZioNi MiNerVa MeDicaOnline version at http://www.minervamedica.it

De Blasi MICROVASCULAR BED CHANGES AFTER PASSIVE LEG RAISING


distensible.7, 17 the relationship between the Plr-induced changes in microvascular Nirs measurements and systemic cardiovascular variables was also assessed.


this observational prospective study was conducted at the sant’andrea University Hos-pital, Rome, Italy, from April 2014 to October 2015 and enrolled 40 critically ill patients (age >18 years) admitted to the general icU with any of the following criteria: systemic inflam-matory response syndrome,9 at least one or-gan requiring functional support, a Simplified acute Physiology score (saPs ii) 10 ≥21, and an injury severity score (iss) 11 ≥16 for trauma patients (group a). the exclusion criteria were: cardiopulmonary edema, severe-to-moderate valvular disease, a history of severe arterial hypertension, acute liver failure, hemorrhagic shock, chronic dialysis, brain death, patients supposed to be discharged from the icU within 24 hours of presentation and those with a high Body Mass Index (>29.9 kg/m2). We enrolled 30 subjects as healthy controls (group B) in whom we monitored hemodynamic variables using non-invasive devices. the study was approved by the institutional research board (Ref.: 7173/2013) and all subjects or their next of kin gave informed consent to the anonymous use of their personal data at hospital admission.

Protocol

all subjects were placed on a reclining bed (totalcare® P500, Hill-Rom Corp., Batesville, IN, USA) in a semi-recumbent position with the trunk raised to 45° relative to the lower limbs. a pneumatic cuff was placed around the arm and connected to an automatic infla-tion system (Hokanson Rapid Cuff Inflator and AG101 Air Source, PMS Instruments Ltd, Maidenhead, England, UK) capable of reaching a predefined cuff pressure (Pcuff) in less than 0.5 seconds. in each subject a Nirs probe was positioned on the upper face of the brachioradial muscle. to avoid differences in hydrostatic pressure requiring a correction fac-

Theoretically, blood volume recruited by PLR or fluid loading enlarges stressed volume increasing cardiac preload through a rise in the mean systemic filling pressure that becomes rapidly apparent and of short duration despite prolonged leg elevation.4 Despite the belief that unstressed volume (Vu) has only a lim-ited involvement in blood volume expansion, particularly in the body part not affected by changes in gravity, only measurements derived from changes in macrocirculation and indirect data support this concept. about 80% of blood volume is contained in veins and three-quar-ters of that in small veins and venules.5, 6 al-though the vascular response to fluid loading primarily involves the microvascular bed only a small number of studies have investigated the venous compartment of microcirculation in humans 7 and no study has investigated how rapid changes in blood volume influence Vu and stressed volume (Vs) and pressure therein.

Equally importantly, no information has yet shown how the venular compartment reacts to Plr in patients admitted to an intensive care unit (ICU) with a systemic inflammatory re-sponse syndrome — a condition which is likely to cause microcirculatory dysfunction. Nor do we know whether eventual changes in the mi-crovascular bed correlate with cardiac respons-es to fluid loading. Answering these questions might help when deciding whether critically ill patients should undergo fluid loading.

In this observational study, we aimed to highlight how Plr-induced blood volume re-cruitment distributes in the skeletal muscle mi-crovascular bed of critically ill patients, possi-bly modifying pressures therein. To do so, first we measured Plr-induced changes in the mi-crovascular bed volume (MBV), its partition-ing in stressed and unstressed volumes, the in-side pressures, tissue blood flow (tBF), elastic compliance of microvascular bed (mvec) and venular tone (tv) using quantitative near-infra-red spectroscopy (NIRS). Then, we compared the values of these variables to those of healthy subjects. the microvascular bed we investi-gated comprised mainly venules, small veins and to a lesser extent, capillaries, since arte-rial vessels account for only 3% and are poorly

MICROVASCULAR BED CHANGES AFTER PASSIVE LEG RAISING De Blasi


Near-infrared spectroscopy settings and mea-surements

Skeletal muscle oxyHb/Mb and deoxyHb/Mb concentrations in μM ([HbO2/MbO2][HHb/Mb]) were measured using a NIMO-4 continuous-wave quantitative photometer 15 (Nirox srl, Brescia, Italy). Data were acquired at a sampling time of 1 second, thus yielding adequate data to analyze changes in tissue Hb concentration. a detailed description of the methods used to calculate microvascular vol-ume variables (MBV, Vu and Vs), pressures in the MBV (PV), in the Vu (threshold pressure: Pit, equal to pressure outside vessels) and in the Vs (Ps) is provided in the online supple-mentary Materials, together with the methods used to calculate mvec, Tv and tBF. Basically, to obtain these variables we applied the same method as used for strain-gauge plethysmog-raphy 16 adding the Nirs measurements of MBV and its changes during cumulative Pcuff increases in steps from 5 mmHg and increas-ing pressure in 10, 15, 20, 25, 30, 40 and 50 mmHg 7 (Figure 1). We limited cuff occlusion

tor, we kept the arm cuff and NIRS probe at the same level by containing the forearm in a plastic frame keeping the arm and the elbow at an angle of 135° at the heart level. For PLR we tilted the entire bed rapidly lowering the sub-ject’s trunk to the supine position and raising the lower part of the bed to 45°.4

For each subject, the study protocol included a first baseline set of NIRS measurements after subjects had rested for about 15 minutes. about 90 seconds after the PLR manoeuver, NIRS variables were measured again and compared with baseline measurements. We calculated the percentage maximum change for each hemo-dynamic variable from baseline to Plr and recorded the time needed for co and stroke volume (SV) to increase by ≥10% 4, 12 (respond-ers), the time at which the increase took place, and the duration of the maximum change.

Invasive hemodynamic monitoring

For measuring central venous pressure (CVP), considered a marker of PLR-induced blood volume recruitment,13 all subjects in group a had a 3-lumen central venous catheter inserted into the internal jugular veins and con-nected to a disposable pressure transducer. Mean arterial pressure (MaP) and co were continu-ously monitored using a thermistor device close to the central venous catheter and a thermistor-tipped arterial catheter inserted into the femoral artery both connected to an eV 1000 clinical platform monitor (edwards lifesciences corp. Irvine, CA 92614, US). After a calibration of the system performed by thermodilution before PLR, we registered changes in CO or SV based on the arterial pulse contour analysis through-out the study (Volume View set).

Finger arterial pressure CO monitoring

in group B non-invasive arterial blood pres-sure (aBPNi) and co (coNi) were measured using a photoplethysmograph system based on a volume clamp method (Nexfin technology, BMEYE B.V, Amsterdam, The Netherlands) using an appropriately sized cuff applied to the mid-phalanx of the middle finger on one hand.14

Figure 1.—Representative recording of microvascular blood volume (MBV) measured in healthy subjects with Nirs at various cuff occlusion pressures (Pcuff). linear blood volume change in the range of 20-50 mmHg was taken for calculat-ing microvascular compliance. Linear regression for the first two MBV changes extrapolated to 0 yielded intravascular threshold pressure (Pit), the intravascular pressure (Pi) cor-responding to extravascular pressure (Pext). the Pcuff-induced MBV changes between the first raise and linear increase represent microvascular bed recruitment. Unstressed vol-ume (Vu) was calculated measuring the MBV at Pit whereas stressed volume (Vs) was derived subtracting Vu from the MBV measured before venous occlusion.



and 15 of them were in septic shock (37.5%).9 Norepinephrine (Ne) was the only vasoactive agent used in the septic shock patients and in two patients without sepsis. Data of macro and microvascular variables in the subgroups of patients in group a are reported online (sup-plementary table i).

Effects of PLR on systemic cardiovascular variables

Responses in CO and/or SV increase to PLR differed markedly in the two groups.

all subjects apart from four in group B (87%) showed a significant CO or SV in-crease, compared to only 15 of the 40 patients in group A (37.5%). In this group, the distri-bution of subjects who showed Plr-induced cardiovascular changes (responders) varied markedly among the subgroups of patients: the number of responders was greater in pa-tients with the only sepsis (8 out of 11, 73%) than those with septic shock (3 out of 15, 20%; P=0.007) or in patients without signs of sepsis (4 out of 14, 29%; P=0.040) (Online Supple-mentary Materials). in the 15 responders in group A, the increase in CO or SV appeared within 30 seconds and lasted an average of 62 seconds (range 60-105 seconds). In group B, the time for the co or sV increase to appear approached that in group a but the increase lasted longer than in group A (mean 112 s, range 94-121 s; P=0.024). Baseline MAP and sV values were lower in group a than in group B but Plr reduced heart rate (Hr) and MaP only in group B (Table II). After PLR, CVP increased in 30 of 40 patients (75%), remained constant in 6, and decreased in 4, but the in-crease lasted an average of 70 seconds (range 60-95 s). No correlation was found between cVP and Pit or Ps values, or between the PLR-induced changes in co and Nirs microvascu-lar variables in both groups.

Effects of PLR on microvascular volume and pressure variables

Baseline MBV, Vu and Vs values for group a matched those for group B whereas the

pressure to a value that was 10 mmHg lower than the individual patient’s diastolic pressure, thus affecting veins rather than arteries.17

Because the Mb concentration in tissue re-mains constant throughout the study, we at-tributed changes in light absorption after ve-nous occlusion only to [HbO2] and [HHb]. tv

18 was calculated from the Plr-induced changes in MBV after the microvascular bed was fully recruited and, because the higher the MBV change, the lower was the vascular tone, we expressed the opposite values in mL as a measure of tone, whereas the mvec was calculated from the slope of the MBV increase measured at the two higher cuff pressures.7, 19 Hence, these variables reflect the elastic and tone properties of the stressed volume.


We calculated that a sample size of 40 pa-tients would have 1.0 power to detect a mini-mum 0.7 units of mean difference, with a stan-dard deviation (sD) of 1.5 for Nirs variables. Kolmogorov-smirnov’s test was used to assess normal data distribution. We used one-way analysis of variance (aNoVa) to test data if normally distributed, otherwise the Mann-Whitney U test for between-group comparison and the Wilcoxon signed rank test for paired data were used. correlation between variables was tested using Pearson or spearman’s test. categorical variables were compared using the χ2 test. Data were expressed as mean ± sD for aNoVa and median with a 25-75% inter-quartile range (iQr) for the signed rank tests. P values less than 0.05 were considered sta-tistically significant. Data were analyzed using IBM SPSS Statistics software, version 22.0 (IBM Corp., Armonk, NY, USA).

Results

all subjects completed the study. No dif-ference between the two groups was observed in gender, demographic, and anthropometric data, apart from the hemoglobin concentra-tion in the blood, as expected (Table I). In the group A, 26 subjects had signs of sepsis (65%)



the subgroups of patients with and without sep-sis. only in patients with septic shock were the Plr-induced changes in microvascular bed volumes combined with changes in the inside pressures (Vs vs. Ps: r=0.80, P=0.005; Vu vs. Pit: r=0.70, P=0.024). In septic shock, changes in pressures correlated also with changes in mvec (Pit vs. mvec: r=0.75, P=0.013; Ps vs. mvec: r=0.70, P=0.023).

Microvascular tone, compliance and blood flow

Because of the great variability in the data (variation coefficient: 1.63), tests showed no difference in baseline mvec values between the two groups. After PLR, mvec decreased

Plr-induced changes in volume variables distinctly differed between the two groups. After PLR, MBV and Vs increased overall in group a but remained unchanged in group B (table ii). tests used to disclose separately the Plr-induced changes in the Vu and Vs showed that the Vs increased from 0.22 (0.10-0.28) mL/100 mL tissue to 0.34 (0.23-0.66) mL/100 mL tissue in patients without sepsis (P=0.039), whereas the Vu increased from 4.20 (3.01-5.82) mL/100 mL tissue to 5.32 (4.01-11.50) mL/100 mL tissue (P=0.036) in patients with septic shock (Figure 2).

At baseline, NIRS disclosed similar PV, Pit or Ps values in groups a and B. Plr induced no significant changes in microvascular pres-sure variables in both groups (table ii) or in

Table I.—�Demographic, anthropometric and clinical data, diagnosis, severity scores and outcomes of the 40 criti-cally ill patients (group A) and the 30 healthy controls (group B).

Variables group a(N.=40)

group B(N.=30) P value

Sex (M/F), N. 22/18 14/16 0.941Age, mean (SD), years 55.3 (17.1) 46.6 (11.5) 0.092BMI, mean (SD), kg/m2 23.2 (4.5) 22.3 (1.6) 0.552SAPSII, mean (SD) 47.3 (12.8)SOFA, mean (SD) 7.8 (4.5)Hemoglobin, mean (SD), g/100 mL 9.7 (1.0) 13.7 (0.6) <0.001Lactates, mmol/L 2.9 (2.6)pH, units 7.34 (0.04)arterial pco2, mmHg 45.0 (10.1)type of admission

Trauma, N. (%) 10 (25)Medical, N. (%) 16 (40)Unscheduled surgery, N. (%) 14 (35)

clinical diagnosisSevere sepsis, N. (%) 11 (27.5)Septic shock, N. (%) 15 (37.5)Acute respiratory failure *, N. (%) 24 (60.0)intracranial hemorrhage †, N. (%) 5 (12.5)Brain damage †, N. (%) 9 (22.5)acute kidney injury §, N. (%) 7 (17.5)

causes of sepsisPneumonia, N. 24Bacteremia, N. 8Peritonitis, N. 10Urinary tract infection, N. 9Clinical sepsis, N. 2

Mechanical ventilation, N. (%) 30 (75.0)Dose of norepinephrine, median [25-75% IQR], μg/kg/min

0.20 [0.09-0.38]

icU outcome N. (%) 11/29 (27.5)Hospital outcome (died/discharged), N. (%) 5/24 (17.2)BMI: Body Mass Index (Du Bois formula); SAPS: Simplified Acute Physiology Score; SOFA: sequential organ failure assessment.*Acute respiratory failure requiring mechanical ventilation; † Glasgow Coma Score ≤9; § acute kidney injury requiring renal replacement therapy.



subgroups of patients in group A separately, mvec decreased only in patients without sep-sis (0.23 (0.07-0.23) mL/mmHg/100 mL tissue vs. 0.17 (0.07-0.20) mL/mmHg/100 mL tis-sue, P=0.006). Although PLR did not induced statistical changes in tBF the values in septic shock were lower than those measured in the other subgroups of patients and in group B.

the Plr-induced changes in MBV were closely correlated with changes in mvec (r=0.90, P<0.001) and Tv (r=0.72, P<0.001) in group a and only with changes in tv (r=0.77, P=0.002) in group B. In both groups, a close correlation was found between the Plr-induced changes in Vu and mvec (group a: r=0.82, P<0.001; group B: r=0.72, P=0.008) or tv (group A: r=0.80, P<0.001; group B: r=0.83, P<0.001). Among critically ill patients, only in those with only sepsis were the Plr-induced changes in tBF inversely correlated

overall in group a but remained unchanged in the group B (table ii). When we tested the

Table II.—�Hemodynamic and microvascular variables in the 40 critically ill patients (group A) and the 30 healthy controls (group B). Results are presented as variables values before (baseline) and 90 s after passive leg raising (PLR).

group a group BBaseline

a vs.baseline

B

Baseline Prl P value Baseline Prl P value P values

Hemodynamic variablesHR, bpm 95.7 (22.0) 96.5 (21.9) 0.260 71.8 (8.3) 64.4 (7.2) 0.035* 0.133MAP, mmHg 78.9 (12.3) 76.4 (14.3) 0.743 95.8 (9.1) 84.9 (8.7) 0.009* 0.018*CVP, mmHg 7.4 (4.2) 11.0 (4.8) 0.006* – – – –CO, L/min 6.5 (1.3) 6.7 (1.9) 0.069 6.8 (2.1) 7.9 (1.9) 0.022* 0.450SV, mL 68.2 (20.0) 69.41 (23.2) 0.828 96.9 (18.6) 112.1 (19.3) 0.012* 0.005*

Microvascular variablesVolume variables

MBV, mL/100 mL tissue 4.9 [3.2-6.5] 5.7 [3.9-8.1] 0.005* 7.5 (5.3) 7.1 (5.9) 0.330 0.561Vu, mL/100 mL tissue 4.8 [3.1-6.2] 5.2 [3.1-7.7) 0.064 7.2 (5.1) 6.5 (5.8) 0.107 0.493Vs, mL/100 mL tissue 0.2 [0.1-0.3] 0.3 [0.2-5.2] 0.001* 0.3 [0.1-0.5] 0.2 [0.1-0.8] 0.273 0.438

Pressure variables Pv, mmHg 4.9 [4.8-9.7] 7.6 (4.9) 0.151 4.9 [4.8-9.7] 6.9 [4.8-9.6] 0.424 1.000Pit, mmHg 3.0 (4.5) 4.6 (6.3) 0.177 3.8 (2.8) 5.5 (2.9) 0.120 0.441Ps, mmHg 4.4 (3.5) 4.1 (3.5) 0.748 2.7 (2.0) 1.9 (1.3) 0.291 0.109

Microvascular blood flowmBF, mL/min/100 mL tissue 27.4 (42.7) 35.1 (57.5) 0.685 25.6 (22.5) 22.3 (25.0) 0.716 0.678

compliance and tonemvec, mL/mmHg/100 mL tissue

0.11 [0.08-0.28] 0.09 [0.06-0.24] 0.004* 0.21 [0.06-0.33] 0.20 [0.06-0.05] 0.840 0.802

tv, mL/100 mL tissue – 4.2 [-3.5-9.7] – 2.3 [-1.0-8.6] 0.452Data are shown as mean (SD) or median [25-75% IQR].HR: heart rate; MAP: mean arterial pressure; CVP: central venous pressure; CO: cardiac output; Sv: stroke volume; MBV: microvascular blood volume; Vu: unstressed volume; Vs: stressed volume; Pv: venular pressure; Pit: intravascular threshold pressure; Ps: stressed pressure; mBF: microvascular blood flow; mvec: microvascular elastic compliance; Tv: venular tone.*P<0.05.

Figure 2.—Unstressed (black bar) and stressed (white bar) volumes at baseline and after passive leg raising in critically ill patients with and without sepsis and septic shock.† stressed volume vs. baseline, P=0.039; † unstressed vol-ume vs. baseline, P=0.036.



although Plr caused a co or sV increase in 37.5% of our critically ill patients, the in-crease in the cVP values proves that this manoeuver recruited blood volume from the lower limbs that was sufficient to cause he-modynamic changes. the Plr-induced effect on systemic perfusion, CO, or SV changes in our patients matched values reported by oth-ers 21, 22 in the percentage of responsiveness and the time frame.

Finding that changes in the stressed volume in healthy subjects and most of critically ill patients are not connected with changes in the inside pressure inside contradicts a paradigm of physiology applied to macrocirculation: that systemic vascular compliance, hence stressed volume compliance, remains constant after flu-id loading. Based on this assumption, Ander-son 23 proposed a derived estimation of Vs in the upper arm by the blood stop flow procedure and, recently, Maas et al.24 applied this method to patients undergoing cardiovascular surgery for measuring alterations in Vs and Vu follow-ing changes in blood volume. our study show-ing that, basically, in skeletal muscle, changes in blood volume entail changes in microvascu-lar bed volume without concomitant changes in pressure therein highlights that many factors are involved in microcirculation, making the re-lationship between volumes and pressures vari-able and unpredictable. these factors include the recruitment of the microvascular bed,25 changes in the elastic compliance and tone of the vessels, a close interaction between the microvascular bed and the tissue blood flow,26 and, in septic shock patients, a leakage of fluids out of the vascular space. these mechanisms require direct investigation of the microvascu-lar bed and render the use of macrocirculatory variables to derive measurements of stressed and unstressed volumes questionable.

although our study did not intend to de-scribe the venous compartment of microcircu-lation as the only cause of the cardiovascular response to fluid loading, our results could add knowledge on the factors contributing to venous return. Our findings show that, in criti-cally ill patients, some of the fluids following volume expansion fill the microvascular bed

with changes in Vu (r=-0.81; P=0.003) but di-rectly with changes in Vs (r=0.99; P<0.001).

Differences in the NIRS variables between re-sponders and non-responders

tests comparing the Nirs microvascular variables of responders (N.=15) with those of non-responders (N.=25) in group A showed no difference between the values measured before and after Plr. comparing the responders of the two groups, the PLR-induced changes in the Vu showed higher values in group a than in group B (group A, ΔVu 1.5±2.2 mL/100 mL tissue vs. group B, -0.7±1.4 mL/100 mL tissue, P=0.009). Finally, as expected, non-responders in group a showed a higher increase in overall MBV after Plr than responders in group B (group A, ΔMBV 0.4 (-0.1 to 1.8) mL/100 mL tissue vs. group B, -0.3 (-0.8 to 0.0) mL/100 mL tissue, P=0.015).

Discussion

The first new finding in this study is that PLR-induced blood volume recruitment enlarges the microvascular venous compartment of skeletal muscle, leaving pressures unchanged, in the critically ill patients but not in healthy sub-jects. In addition, changes in the microvascu-lar bed volumes entail concurrent variations in the elastic compliance and tone of venular ves-sels. A second finding is that, among critically ill patients, those with septic shock respond to Plr enlarging the unstressed volume and in-creasing pressures within it (hemodynamically inactive) whereas patients without sepsis en-large the stressed volume (hemodynamically active).20 Another relevant finding of our study comes from the relationships between the un-stressed or stressed volumes and tissue blood flow shown in patients with sepsis and to a lesser extent in healthy subjects, which gives the changes in these venous compartments a role in the control of tissue perfusion. Final-ly, as we supposed, changes in microvascular variables such as blood flow and vascular pres-sures cannot be predicted by changes in mac-rocirculatory cardiovascular variables.



vascular bed,26 hence the tissue perfusion. the different behavior in microvascular regulation between sepsis and the other diseases could somehow explain the higher number of re-sponders in patients with sepsis compared with patients without sepsis or septic shock. How-ever, due to the small number of patients we cannot exclude that a large number of patients could display changes in MBV partitioning.

In septic shock, differently from other clini-cal conditions, fluid loading leads to an increase in the unstressed volume as a consequence of the increase in vessel capacitance resulting from active dilation of small veins/venules. Our finding in humans agrees with the results of experimental studies in animal models of canine and porcine endotoxemia.28, 29 the pres-sure increase in the unstressed volume after flu-id loading reflects also an increase in the pres-sure outside the microvascular bed due to loss of fluids to the interstitium. This phenomenon could also affect tBF and partially explain the deep blood flow reduction observed in septic shock patients. This fluid leakage could reduce the volume to the vascular bed, hence to venous return, and also be an obstacle to oxygen diffu-sion from capillaries to cells. Because we used norepinephrine almost exclusively in patients with septic shock, the role of this vasoactive drug in the microvascular bed response to fluid loading remains unclear, since its effects over-lap the effects due to septic shock. the use of NE could have increased the small vein/venule stiffness, thus justifying the close relationship between changes in the microvascular volumes and pressures. Equally, NE could be also in-volved in the tBF reduction and in the absence of a relationship between changes in tBF and the stressed or unstressed volumes impeding to maintain constant the tissue perfusion. How-ever, our study might add valuable information to what others have observed about the effects of norepinephrine on preload and co.30, 31

even though our results disclose changes in MAP and HR only in healthy subjects, we cannot exclude the possibility that Plr could act in both groups by differentially activating microvascular tone through the baroreflex or low-pressure baroreceptors.32, 33

differently from those in healthy subjects in whom neurogenic control possibly makes flu-ids to be more available for systemic circula-tion. Inflammation affecting the neurogenic control of the microvascular bed could cause a reduction in the elastic compliance as a re-sult of the recruitment of vessels and vasodila-tion. the possibility that the microvascular bed could be recruited has been amply documented in skeletal muscle and recently confirmed in a previous in-vivo study on healthy subjects 7 as a consequence of a pressure increase in the outflow. In addition, the decoupling between the microvascular volumes and pressures has clinical relevance because it entails a lack of increase in the post-capillary pressure follow-ing stressed volume enlargement that prevents this blood volume from contributing to venous return. this could be ascribed as a causal factor for the large number of non-responders among critically ill patients. The finding that enlarge-ment of the microvascular bed occurs also in patients who respond to fluid loading could possibly be explained by the shorter cardiovas-cular response that these patients show (62 s) compared with healthy subjects (112 s). Hence, the microvascular bed changes that we detected 90 s after Plr could exceed the short-lasting effects of fluid loading on macrocirculation.

the different behaviors between patients with sepsis and septic shock concerning the microvascular bed partitioning and capaci-tance after fluid loading derives from factors linked to the inflammatory mediators,27 the se-verity of sepsis disease, and the use of the va-soactive drugs. In sepsis, the close correlation between changes in the MBVs and in elastic compliance is probably due to the additive ef-fects of circulating mediators, NO,27 and neu-rogenic control. We hypothesize that, in sep-sis, the increase in elastic compliance could be opposed by the concomitant increase in venu-lar tone likely due to the stretching of vessels resulting in a lack of evident changes in the stressed or unstressed volumes. Because of the close interaction between the tissue blood flow and the stressed or unstressed volume, we first found that in only sepsis fluid loading does not involve changes in the time constant of the



should seek greater insight into the difference in microvascular partitioning among patients with and without sepsis or septic shock, and assess the role of vasoactive drugs in the rela-tionship between the venous compartment of microcirculation and venous return.

Key messages

— in skeletal muscle changes in mi-crovascular bed, stressed and unstressed volume, are uncoupled with changes in the pressures inside. This finding makes ques-tionable use of macrocirculation to derive data on stressed or unstressed volumes.

— PLR and possibly fluid loading lead to an enlargement of the microvascular bed in critically ill patients unlike healthy sub-jects. this different behavior in microvas-cular response implies that fluid loading in subjects with a diffuse inflammation with or without sepsis entails transferring part of the blood volume in microcirculation thus affecting the venous return and the cardio-vascular response to volume expansion.

— a sudden increase in the blood vol-ume enhances the stressed volume in pa-tients without sepsis and the unstressed vol-ume in patients with septic shock. in septic shock fluid loading induces also a loss of fluids to the interstitium thus hindering the oxygen diffusion from capillaries to cells.

— in critically ill patients the respond-ers to fluid loading do not differ from non-responders in the macrovascular bed changes. in these patients the cardiovascu-lar response to fluid loading is shorter than in healthy subjects.

References

1. Boulain T, Achard JM, Teboul JL, Richard C, Perrotin D, ginies g. changes in BP induced by passive leg raising predict response to fluid loading in critically ill patients. Chest 2002;121:1245-52.

2. Geerts BF, Aarts LP, Groeneveld AB, Jansen JR. Predict-ing cardiac output responses to passive leg raising by a PEEP-induced increase in central venous pressure, in car-diac surgery patients. Br J Anaesth 2011;107:150-6.

the relationship between systemic hemo-dynamics and microcirculation changes is controversial.34-37 Although a significant cor-relation between systemic hemodynamic pa-rameters and some microcirculatory indices was shown in the early phase of septic shock in patients at the emergency department,35 it disappeared in the icU patients.34, 35

The deep reduction in the tissue blood flow that we found in septic shock patients results from factors that had occurred over time, thus making microcirculation insensitive to chang-es in macrocirculation.


our study has limitations. one is that the microvascular bed of skeletal muscle contrib-utes about 20% to the overall blood volume.38 this relatively small contribution could make the observed change less meaningful. Despite this drawback, the changes and correlations that we observed imply that the microvascu-lar changes we noticed in response to Plr-induced preload changes could be extended to the microcirculation in other body areas, as other authors support.24 another limitation is the possible inaccuracy of the Mb concentra-tion in tissue; notwithstanding, although the absolute Mb value might be inaccurate, we consider the microvascular bed changes due to Plr manoeuvre in the same subject reliable. a final limitation is the small number of patients in the subgroups studied even though our study yields useful data for clinical purposes.

Conclusions

In summary, our study evidenced that the use of macrocirculatory parameters to derive measurements of the stressed and unstressed volumes is questionable. From our data on patients with septic shock, the PLR-induced enlargement of the unstressed volume induces also loss of fluids to the interstitium, hence clinicians should be aware that fluid loading in these patients could leave organ perfusion unchanged or cause it to worsen. apart from enrolling a larger study sample, future research



ard C, Pinsky MR, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med 2006;34:1402-7.

22. Cherpanath TGV, Geerts BF, Lagrand WK, Schultz MJ, Groeneveld ABJ. Basic concepts of fluid responsiveness Neth Heart J 2013;21:530-6.

23. anderson rM. the gross Physiology of the cardiovas-cular System. Tucson, AZ: Racquet Press; 1993.

24. Maas JJ, Pinsky MR, Aarts LP, Jansen JR. Bedside Assessment of Total Systemic Vascular Compliance, Stressed Volume, and Cardiac Function Curves in Inten-sive Care Unit Patients Anesth Analg 2012;115:880-7.

25. Vallet B. endothelial cell dysfunction and abnormal tis-sue perfusion. Crit Care Med 2002;30:S229-34.

26. Madger S. Shock physiology. In: Pinsky MR, Dhainhault JF, editors. Pathophysiologic foundations of critical care. Baltimore, MD: Williams and Wilkins; 1993. p. 140-60.

27. Stearns-Kurosawa DJ, Osuchowski MF, Valentine C, Ku-rosawa S, Remick DG.The Pathogenesis of sepsis. Annu rev Pathol 2011;6:19-48.

28. Bressack MA, Morton NS, Hortop J. Group B strepto-coccal sepsis in the piglet: Effects of fluid therapy on ve-nous return, organ edema, and organ blood flow. Circ Res 1987;61:659-69.

29. Hiesmayr M, Jansen JR, Versprile A. Effects of endotoxin infusion on mean systemic filling pressure and flow re-sistance to venous return. Pflugers Arch 1996;431:741-7.

30. Persichini R, Silva S, Teboul JL, Jozwiak M, Chemla D, Richard C, et al. effects of norepinephrine on mean sys-temic pressure and venous return in human septic shock. Crit Care Med 2012;40:3146-53.

31. Monnet X, Jabot J, Maizel J, Richard C, Teboul JL. Nore-pinephrine increases cardiac preload and reduces preload dependency assessed by passive leg raising in septic shock patients. Crit Care Med 2011;39:689-94.

32. Pang CCY. Measurement of body venous tone. J Pharma-col Toxicol Methods 2000;44:341-60.

33. Parati G, Grassi G, Coruzzi P, Musiari L, Ravogli A, No-varini A, et al. Influence of cardiopulmonary receptors on the bradycardic responses to carotid baroreceptor stimu-lation in man. Clin Sci (Lond) 1987;72:639-45.

34. De Backer D, Creteur J, Dubois MJ, Sakr Y, Koch M, Verdant C, et al. the effects of dobutamine on micro-circulatory alterations in patients with septic shock are independent of its systemic effects. crit care Med 2006;34:403-8.

35. Ospina-Tascon G, Neves AP, Occhipinti G, Donadello K, Büchele G, Simion D, et al. Effects of fluids on microv-ascular perfusion in patients with severe sepsis. intensive Care Med 2010;36:949-55.

36. Trzeciak S, Dellinger RP, Parrillo JE, Guglielmi M, Ba-jaj J, Abate NL, Arnold RC, et al. early microcirculatory perfusion derangements in patients with severe sepsis and septic shock: relationship to hemodynamics, oxygen transport, and survival. Ann Emerg Med 2007;49:88-98.

37. Jansen TC, van Bommel J, Schoonderbeek FJ, Sleeswijk Visser SJ, van der Klooster JM, Lima AP, et al. early lactate-guided therapy in intensive care unit patients: a multicenter, open-label, randomized controlled trial. Am J Respir Crit Care Med 2010; 182:752-61.

38. Greenway CV, Lautt WW. Blood volume, the venous sys-tem, preload and cardiac output. Can J Physiol Pharmacol 1986;64:383-7.

3. Maizel J, Airapetian N, Lorne E, Tribouilloy C, Massy Z, Slama M. Diagnosis of central hypovolemia by using passive leg raising. Intensive Care Med 2007;33:1133-8.

4. Monnet X, Teboul JL. Passive leg raising. Intensive Care Med 2008;34:659-63.

5. Abboud FM, Mark AL, Heidstadd DD, Schmid PG, Barnes RW. The venous system. In: Levine HJ, editor. clinical cardiovascular physiology. New York: grune and Stratton; 1976. p. 207-57.

6. Greenway CV, Lister GE. Capacitance effects and blood reservoir function in the splanchnic vascular bed during non-hypotensive hemorrhage and blood volume expan-sion in anaesthetized cats. J Physiol 1974;237:279-94.

7. De Blasi RA, Arcioni R. Assessing skeletal muscle vari-ations in microvascular pressure and unstressed blood volume at the bedside. Microcirculation 2014;21:606-14.

8. Mancini DM, Bolinger L, Li H, Krndrick K, Chance B, Wilson JR. Validation of near-infrared spectroscopy in humans. J Appl Physiol 1994;77:2740-7.

9. American College of Chest Physicians/ Society of Criti-cal Care Medicine Consensus Conference. Definitions of sepsis and organ failure and guidelines for the use of inno-vative therapies in sepsis. Crit Care Med 1992;20:864-74

10. Le Gall JR, Lemeshow S, Saulmer F. A new simpli-fied acute physiology score (SAPS II) based on a Eu-ropean/North American multicentre study. JAMA 1993;270:2957-63.

11. Civil ID, Schwab CW. The abbreviated injury scale, 1985 revision: A condensed chart for clinical use. J Trauma 1988;28:87-90.

12. Marik PE, Monnet X, Teboul JL. Hemodynamic parame-ters to guide fluid therapy. Ann Intensive Care 2011;1:1-9.

13. Antonelli M, Levy M, Andrews PJ, Chastre J, Hudson LD Manthous C, et al. Hemodynamic monitoring in shock and implications for management. intensive care Med 2007;33:575-90.

14. Monnet X, Dres M, Ferré A, Le Teuff G, Jozwiak M, Bleibtreu A, et al. Prediction of fluid responsiveness by a continuous non-invasive assessment of arterial pressure in critically ill patients: comparison with four other dy-namic indices. Br J Anaesth 2012;109:330-8.

15. Matcher S, Cope M, Delpy D. Use of the water absorp-tion spectrum to quantify tissue chromophore concentra-tion changes in near-infrared spectroscopy. Phys Med Biol 1994;39:177-96.

16. Gamble J, Christ F, Gartside IB. Mercury in silastic strain gauge plethysmography for the clinical assessment of the microcirculation. Postgrad Med J 1992;68:S25-S33.

17. Rothe CF. Mean circulatory filling pressure: its meaning and measurement. J Appl Physiol 1993;74:499-509

18. Aaronson PI, Ward JPT, Wiener CM. The venous system. In: Aaronson PI, Ward JPT, editors. The cardiovascular system at a glance. Oxford: Wiley-Blackwell; 1999. p. 46-59.

19. De Blasi RA, Palmisani S, Alampi D, Mercieri M, Ro-mano R, Collini S, et al. Microvascular dysfunction and skeletal muscle oxygenation assessed by phase-modu-lation near-infrared spectroscopy in patients with septic shock. Intensive Care Med 2005;31:1661-8.

20. Funk DJ, Jacobsohn E, Kumar A. the role of venous re-turn in critical illness and shock—Part i: Physiology. crit care Med 2013;41:255-62.

21. Monnet X, Rienzo M, Osman D, Anguel N, Rich-

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Manuscript accepted: March 31, 2016. - Manuscript revised: March 1, 2016. - Manuscript received: September 9, 2015.For supplementary materials, please see the online version of this article.


ing of its multiple modes of action and effects on pain modulation and central sensitization,1 ketamine has become a rational option for improving perioperative pain management.

Ketamine, which was first introduced to an-esthesia in the 1970s, has regained broad

interest in perioperative medicine within the last few years. With increasing understand-


the effects of minimal-dose versus low-dose s-ketamine on opioid consumption,

hyperalgesia, and postoperative delirium: a triple-blinded, randomized, active- and

placebo-controlled clinical trialHelmar BorNeMaNN-ciMeNti, Mischa WeJBora, Kristina MicHaeli,

alexander eDler, andreas saNDNer-KiesliNg

Department of anesthesiology and intensive care Medicine, Medical University of graz, graz, austria*corresponding author: Helmar Bornemann-cimenti, Department of anaesthesiology and intensive care Medicine, Division of special anesthesiology, Pain and intensive care Medicine, Medical University of graz, auenbruggerplatz 29, a-8036 graz, austria. e-mail: [email protected]


lavoro: 11019-Mastitolo breve: MiNiMal-Dose VersUs loW-Dose s-KetaMiNeprimo autore: BorNeMaNN-ciMeNtipagine: 1069-76citazione: Minerva anestesiol 2016;82:1069-76

a B s t r a c tBACKGROUND: Evidence confirms that perioperative ketamine administration decreases opioid usage. To reduce the risk for potential psychodysleptic side effects, however, ketamine dosing tends to be limited to low-dose regimens. We hypothesized that even lower doses of ketamine would be sufficient, with minimal side effects, when used as a component of multimodal perioperative pain management.MetHoDs: in this triple-blinded, randomized, active- and placebo-controlled clinical trial, patients undergoing elective major abdominal surgery were randomized to one of three treatment groups: low-dose s-ketamine (a 0.25 mg/kg bolus and 0.125 mg/kg/h infusion for 48 hours), minimal-dose s-ketamine (a 0.015 mg/kg/h infusion following a saline bolus), and placebo (saline bolus and infusion). opioid consumption, pain levels, hyperalgesia at the incision site, and delirium scores were assessed 48 h postoperatively.resUlts: Patients in the placebo group had the highest cumulative piritramide consumption and the largest normalized areas of hyperalgesia at the incisional site, while those in the low-dose group had the highest delirium scores. Postopera-tive pain levels did not differ significantly between the treatment groups.coNclUsioNs: our data demonstrate that minimal-dose s-ketamine was comparable to the conventional low-dose regimen in reducing postoperative opioid consumption and hyperalgesia. Postoperative delirium, however, was less fre-quent with the minimal-dose regimen. We therefore suggest that minimal-dose s-ketamine may be a useful low-risk component of balanced perioperative analgesia.(Cite this article as: Bornemann-cimenti H, Wejbora M, Michaeli K, edler a, sandner-Kiesling a. the effects of minimal-dose versus low-dose s-ketamine on opioid consumption, hyperalgesia, and postoperative delirium: a triple-blinded, rand-omized, active- and placebo-controlled clinical trial. Minerva anestesiol 2016;82:1069-76)Key words: Ketamine - Hyperalgesia - Delirium - Perioperative period - Postoperative pain.


BorNeMaNN-ciMeNti MiNiMal-Dose VERSUS loW-Dose s-KetaMiNe


Inclusion and exclusion criteria

inclusion criteria were elective major open abdominal surgery (colorectal and hepatic sur-gery), age >18 years, weight between 40 and 120 kg, and american society of anesthesiolo-gists (asa) physical status class i-iii. exclu-sion criteria were any acute or chronic pain state treated with opioid therapy, severe liver or kid-ney dysfunction, severe coronary disease, preg-nancy, addiction to alcohol or opioids, present or past psychotic disorders, and poor compliance.

Study groups

after receiving written consent, patients were equally randomized to one of three treat-ment regimens, as follows:

— low-dose group: a 0.25 mg/kg intrave-nous (i.v.) bolus of s-ketamine after induction of anesthesia followed by a 0.125 mg/kg/h continuous i.v. infusion of s-ketamine for 48 hours;

— Minimal-dose group: a 0.9% i.v. saline bolus after induction of anesthesia followed by a 0.015 mg/kg/h continuous i.v. infusion of s-ketamine for 48 hours;

— Placebo group: a 0.9% i.v. saline bolus after induction of anesthesia followed by a 0.9% i.v. saline infusion for 48 hours.

the low-dose regimen was administrated following the protocol published by De Kock et al.8 S-ketamine was purchased from Pfiz-er (Ketanest S; Pfizer Cooperation Austria gmbH, Vienna, austria).

Randomization and blinding

the study medication was prepared preopera-tively, using a computer-generated randomization list, by an anesthetist who was excluded from any further involvement in the treatment or evaluation of patients. the two syringes for the bolus injec-tion and the continuous infusion were labeled with “study Medication” and the randomization number of the patient. all involved patients, nurs-es, and physicians were strictly blinded to group allocation. likewise, the statistician remained blinded until the end of the analysis.

Indeed, a broad evidence base now confirms that the perioperative administration of ket-amine may decrease opioid consumption.2-5 to reduce the risk for potential psychodysleptic side effects, however, the administered doses of ketamine have been limited to a maximum infusion rate of 1.2 mg/kg/h or an intraopera-tive bolus of 1 mg/kg, which are considered low-dose regimens.6

a recent meta-analysis of different low-dose ketamine regimens showed considerable variation in the modes of application and doses used.6 this was also observed in a recent French survey.7 in the current literature, some studies have used a single preoperative bolus, some have administrated ketamine by continuous in-fusion, and some have applied a combination of a bolus injection with continuous infusion. Moreover, dosages have varied considerably in this research. Despite these limitations, con-tinuous low-dose ketamine administration for 24-48 hours has been associated with superior effects on residual pain when compared with a single bolus administration, but without major complications.6 No clear dose-effect relation-ship has been detected to date.

We hypothesize that when used as part of the multimodal treatment of perioperative pain, not only would good relief be achieved by low-er doses of ketamine than are currently used in low-dose regimens, but that this approach would also be associated with a lower risk for psychodysleptic and other side effects.6 thus, we compared the effects of minimal- and low-dose ketamine regimens on opioid consump-tion, hyperalgesia at the incisional site, and postoperative delirium.


Study design

this study was conducted as a triple-blind-ed, randomized, active- and placebo-con-trolled clinical trial at the Medical University of Graz, Austria (ClinicalTrials.gov Identifier: Nct01022840). after receiving ethics com-mittee approval, eligible patients were con-tacted and informed about the study.

MiNiMal-Dose VERSUS loW-Dose s-KetaMiNe BorNeMaNN-ciMeNti


sia.10 this psychometric tool gave a maximum score of eight points from eight different di-mensions (level of consciousness, inattention, disorientation, hallucination–delusion–psy-chosis, psychomotor agitation or retardation, inappropriate speech or mood, sleep/wake cycle disturbance, and symptom fluctuation). a score of more than three was taken to in-dicate delirium.10 the german version has previously been validated in a postoperative population.11

Sample size calculation

When designing the study, no data were available to estimate the effect of minimal-dose s-ketamine on hyperalgesia or postop-erative opioid consumption. We therefore calculated the a priori sample size assum-ing the differences between the dose groups and the standard deviation of 50% each. setting alpha to 0.05 and beta to 20%, we calculated that an appropriate group size would require 16 patients. We planned to include 20 patients per group (N.=60) to al-low for potential dropouts or protocol vio-lations.


Unless otherwise specified, data are ex-pressed as means ± standard deviations or as medians (95% confidence intervals). Data were tested for normality by using the shap-iro-Wilk’s W test. Differences between groups were calculated by analysis of variance or the Kruskal-Wallis test, as appropriate. the low- and minimal-dose groups were further com-pared by student’s t-tests or Mann–Whitney U tests, as appropriate. as this procedure could be interpreted as a multiple comparison, the Bonferroni correction was applied. statistical analysis was performed using Ncss 10 sta-tistical software (Ncss, llc., Kaysville, Ut, Usa).

this manuscript was prepared according to the consolidated standards of reporting tri-als (coNsort) statement. 12.

Study protocol

Patients were introduced preoperatively to an 11-point numerical rating scale (Nrs) for pain assessment and to a patient controlled analgesia (Pca) device. anesthesia was in-duced and maintained following our insti-tutional standards: premedication with 7.5 mg oral midazolam; induction with 2 µg/kg fentanyl, 2-3 mg/kg propofol, and 0.6 mg/kg rocuronium; balanced anesthesia with 0.7-1.0 Mac (minimum alveolar concentration) sevoflurane and 0.1-0.3 µg/kg/min remifen-tanil; and, for postoperative analgesia, 0.2 mg/kg piritramide and 75 mg i.v. diclofenac, given approximately 30 minutes before the end of surgery. No local or regional anes-thesia was applied. after recovery, Pca was provided with i.v. piritramide (caDD®-solis; smith Medical asD inc., st. Paul, MN, Usa) set to a bolus size of 0.02 mg/kg, a lock-out time of 10 minutes, and a maxi-mum of five boluses per hour. In addition, 75 mg diclofenac was administered twice daily by infusion.

Hyperalgesia was evaluated 48 h postop-eratively, immediately after the end of the continuous infusion, using a method pre-viously described by our group.9 Briefly, a 180-mN von-Frey filament (Semmes–Wein-stein Monofilament; Touch-Test™ Sensory evaluators, North coast Medical inc., Mor-gan Hill, ca, Usa) was used to test areas around the wound from outside to the cen-ter of the incision along four radial lines in 5-mm steps. The first spot on the skin where the patient experienced a “sharp,” “sore,” or “painful” sensation was marked and the distance to the incision recorded. the data were normalized to exclude the effect of in-cisional length on the area of hyperalgesia.9 Pain levels were assessed at rest and during movement using the 11-point Nrs every 4 hours. in addition, piritramide use was docu-mented 48 h postoperatively.

Postoperative delirium was assessed us-ing the intensive care Delirium screening checklist (icDsc) 48 hours postoperatively, immediately after the evaluation of hyperalge-



sion were postoperative intubation (N.=2), withdrawal of consent (N.=1), and reoperation during the study period (N.=1). the patient flow chart is presented in Figure 1, and the pa-tients’ characteristics are presented in table i.

Results

We included 60 patients over a 42-month period, but 4 patients were excluded, leaving 56 for the final analysis. Reasons for exclu-

Figure 1.—Flow chart of the study.

Exclusionnot meeting inclusion criteria (N.=76)

declining to participate (N.=33)

Enrollment

Assessment for eligibility (N.=169)

Randomization(N.=60)

Allocation

Minimal-dose group(N.=20)

received allocated intervention (N.=20)

Low-dose group(N.=20)


Placebo group(N.=20)


Follow-up Follow-up Follow-up

Patient remained intubated (N.=1) Patient remained intubated (N.=1)Patient withdraw consernt (N.=1) Patient underwent reoperation (N.=1)

Analysis Analysis Analysis

Analyzed (N.=19) Analyzed (N.=18) Analyzed (N.=19)



dose groups compared with the placebo group (low: 42.7±13.4; minimal: 40.2±13.5; pla-cebo: 72.7±15.3; P<0.0001), it did not differ between the two treatment groups (P=0.5850).

the details of piritramide use and pain lev-els (at rest and during movement) over time are presented in Figures 2 and 3, respectively. No significant differences were observed be-tween the pain scores during rest or movement at any point. However, the icDsc score was significantly increased in the low-dose group

Because all patients were treated as defined in the protocol, we did not perform an intention-to-treat analysis.

There were no significant differences in age, weight, height, or ASA classification be-tween the groups. Likewise, no significant dif-ferences were found in the length of surgery (skin incision to closure) or in the intraopera-tive remifentanil dose between the groups. al-though the cumulative piritramide dose was significantly lower in the low- and minimal-

Figure 2.—Postoperative piritramid consumption over time. Piritramid consumption at 4-hour intervals in the 48 h postop-eratively. Patients are separated by treatment group (low-dose group: 0.25 mg/kg s-ketamine bolus plus a 0.125 mg/kg/h continuous infusion of s-ketamine for 48 h; minimal-dose group: 0.015 mg/kg/h i.v. infusion of s-ketamine for 48 h; and placebo group: 0.9% i.v. saline infusion for 48 h).Data are reported as mean ± standard error.*P<0.05 in the analysis of variance; § P<0.05 when comparing the minimal- and low-dose groups.

Table I.—�Descriptive data of the patients separated by treatment group (low-dose group: S-ketamine bolus of 0.25 mg/kg followed by S-ketamine continuous infusion of 0.125 mg/kg/h for 48 hours, minimal-dose group: i.v. S-ketamine infusion of 0.015 mg/kg/h for 48 hours, placebo group: i.v. saline 0.9% infusion for 48 hours).

low dose Minimal dose Placebo P

Height (cm) 170.3±8.2 164.0±21.9 171.6±8.3 0.2238Weight (kg) 78.0±11.6 75.3±13.8 75.4±10.8 0.7390sex (F/M) 9/9 11/8 11/8 0.8573asa (1/2/3/4) 1/5/12/0 0/11/8/0 3/8/8/0 0.1444age (years) 62.2±9.8 58.4±8.1 61.0±12.4 0.4256Duration of surgery (min) 183±72 182±71 189±67 0.7948remifentanil consumption (µg) 3298±1171 3393±1226 3430±1385 0.9311Data are reported as mean ± sD.



lative consumption at 48 hours being 40% lower in both treatment groups compared with the placebo group. these data are con-sistent with those of previous reports,6 and show comparability between conventional low-dose treatment and minimal-dose treat-ment. Significantly, however, postoperative delirium was less frequent with the minimal-dose regimen.

in recent years, multiple modes of action have been suggested to explain the analge-sic properties of ketamine. although some research has shown the effects of ketamine to be dose dependent, others have indicated that this was not the case,1 which is consistent with clinical experience in chronic pain man-agement. Of specific interest to this research, oral ketamine has been shown to enhance analgesic treatment in various diseases 13-15 with oral doses of 1-2 mg/kg/24h. consider-ing that the oral bioavailability of ketamine is about 20%, these doses are surprisingly low compared with those currently used

(median, 2 [1-3]) compared with the minimal-dose (median, 1 [0-1]) and the placebo (medi-an, 0 [0-1]) groups (P=0.007). the normalized area of hyperalgesia was significantly larger in the placebo group (115.9±122.1) than in the two treatment groups (low: 43.9±32.1; mini-mal: 54.2±58.7) (P=0.0048), but did not differ significantly between the two treatment groups (P=0.5763).

Discussion

our data showed that minimal-dose s-ketamine was comparably effective to con-ventional low-dose s-ketamine in reducing postoperative hyperalgesia and opioid con-sumption. Moreover, pain scores did not dif-fer significantly between the three groups at any observed time point. This may reflect proper usage of Pca, because all patients were instructed to aim for satisfactory pain control. Piritramide use varied significantly at every observed time point, with the cumu-

Figure 3.—Postoperative pain levels over time. Pain levels in rest and movement during the 48 h postoperatively separated by treatment group (low-dose group: 0.25 mg/kg s-ketamine bolus plus a 0.125 mg/kg/h continuous infusion of s-ketamine for 48 h; minimal-dose group: 0.015 mg/kg/h i.v. infusion of s-ketamine for 48 h; and placebo group: 0.9% i.v. saline infu-sion for 48 h).Data are reported as mean ± standard error. No significant differences were found at any time point.

0

1

2

3

4

5

0-4h 4-8h 8-12h 12-16h 16-20h 20-24h 24-28h 28-32h 32-36h 36-40h 40-44h 44-48hpostoperative time

low dose - pain in rest

minimal dose - pain in rest

placebo - pain in rest

low dose - pain in movement

minimal dose - pain in movement

placebo - pain in movement

NRS

pain

ratin

g (0

-10)




the major limitation of our study is that we did not provide longer-term data on the development of persistent postoperative pain. although this is doubtlessly a pivotal outcome parameter, we only aimed to demonstrate the clinical effectiveness of a minimal-dose regime in the perioperative peri-od. a long-term follow-up study would represent an interesting topic for further research. High dos-es of intraoperative opioids, particularly remifen-tanil, have been shown to increase postoperative hyperalgesia.30 in our population, however, no significant differences were found in intraopera-tive remifentanil consumption between groups. therefore, we conclude that opioid-induced hy-peralgesia did not skew our observations.

Conclusions

in conclusion, a minimal-dose s-ketamine reg-imen is effective in reducing postoperative opioid consumption and hyperalgesia following abdom-inal surgery and can reduce the risk of postop-erative delirium when compared with a standard low-dose regimen. We therefore suggest that minimal-dose s-ketamine be used as a low-risk component of balanced perioperative analgesia.

perioperatively, and are comparable to those used in our minimal-dose regimen. over the history of perioperative care, we have seen ketamine doses decrease from anesthetic to analgesic doses, and further decrease to the adoption of low-dose regimens comparable to that used in this trial. our data indicated that even lower doses may be sufficient for optimizing pain management among patients after surgery.

Decreasing doses can potentially mini-mize side effects. Postoperative delirium is a major concern in perioperative medicine, and ketamine has been associated with a prolonged time to resumption of mental ori-entation.16 Psychodysleptic symptoms are a well-described side effect of ketamine,1, 17 with a recent meta-analysis confirming neu-ropsychiatric side effects to be significantly associated with perioperative ketamine use.5 indeed, these side effects have been shown to occur with plasma concentrations as low as 50 ng/ml, which roughly equates to a 0.1 mg/kg i.v. dose.18, 19 even though the risk is generally regarded as small, the conscious patient should be informed and monitored carefully.2 By contrast, it is also posited that ketamine could decrease postoperative de-lirium by reducing opioid side effects.20 in our population, the risk for postoperative delirium was highest in the low-dose group, whereas there were no significant differ-ences between the minimal-dose and placebo groups. this is in accordance with the cur-rent literature indicating that psychomimetic effects are dose dependent.1, 17

Hyperalgesia is generally regarded as a consequence of central sensitization and plas-ticity, and has repeatedly been linked to the development of chronic persistent pain.21-25 consequently, its use as a primary parameter is increasingly recommended in studies of in-cisional pain.26-29 Our findings show that there was no significant difference between the low- and minimal-dose ketamine regimens in the reduction of hyperalgesia at the wound site. compared with placebo, the normalized area of hyperalgesia was less than half the size of that in both ketamine treatment groups.

Key messages

— this study compared placebo with the effects of two perioperative s-ketamine dosing strategies on postoperative pain, hyperalgesia, opioid consumption, and delirium.

— Minimal-dose s-ketamine (no start bo-lus plus 0.015 mg/kg/h for 48 hours) was as effective as conventional low-dose s-ketamine (0.25 mg/kg start bolus plus 0.125 mg/kg/h for 48 hours) in reducing postoperative opioid consumption and hyperalgesia, while showing comparable pain control.

— importantly, delirium was found to occur more often in the low-dose group than in either the minimal-dose group or the placebo group.

— We therefore suggest that minimal-dose s-ketamine infusion may be a useful low-risk supplement for balanced perioperative analgesia.



for the relief of intractable chronic pain: a retrospective 5-year study of 51 patients. eur J Pain 2015;19:984-93.

16. chen JY, Jia Je, liu tJ, Qin MJ, li WX. comparison of the effects of dexmedetomidine, ketamine, and placebo on emergence agitation after strabismus surgery in chil-dren. can J anaesth 2013;60:385-92.

17. White PF, Ham J, Way Wl, trevor aJ. Pharmacology of ketamine isomers in surgical patients. anesthesiology 1980;52:231-9.

18. Bowdle ta, radant aD, cowley Ds, Kharasch eD, strassman rJ, roy-Byrne PP. Psychedelic effects of ket-amine in healthy volunteers: relationship to steady-state plasma concentrations. anesthesiology 1998;88:82-8.

19. Hartvig P, Valtysson J, lindner KJ, Kristensen J, Karlsten r, gustafsson ll, et al. central nervous system effects of subdissociative doses of (s)-ketamine are related to plasma and brain concentrations measured with positron emission tomography in healthy volunteers. clin Phar-macol ther 1995;58:165-73.

20. orena eF, King aB, Hughes cg. the role of anesthesia in the prevention of postoperative delirium: a systematic review. Minerva anestesiol. 2016;82:669-83.

21. Brennan tJ, Kehlet H. Preventive analgesia to reduce wound hyperalgesia and persistent postsurgical pain: not an easy path. anesthesiology 2005;103:681-3.

22. Kawamata M, takahashi t, Kozuka Y, Nawa Y, Nishi-kawa K, Narimatsu e, et al. experimental incision-in-duced pain in human skin: effects of systemic lidocaine on flare formation and hyperalgesia. Pain. 2002;100:77-89.

23. Kawamata M, Watanabe H, Nishikawa K, takahashi t, Kozuka Y, Kawamata t, et al. Different mechanisms of development and maintenance of experimental incision-induced hyperalgesia in human skin. anesthesiology 2002;97:550-9.

24. gerbershagen HJ. transition from acute to chronic post-surgical pain. Physiology, risk factors and prevention. schmerz 2013;27:81-93.

25. Wilder-smith oH. chronic pain and surgery: a review of new insights from sensory testing. J Pain Pall care Phar-macother 2011;25:146-59.

26. Brennan tJ, Zahn PK, Pogatzki-Zahn eM. Mechanisms of incisional pain. anesthesiol clin N am 2005;23:1-20.

27. lavand’homme P. Perioperative pain. curr opin anaes-thesiol 2006;19:556-61.

28. Pogatzki-Zahn eM, Zahn PK, Brennan tJ. Postoperative pain--clinical implications of basic research. Best Pract res clin anaesthesiol 2007;21:3-13.

29. lederer aJ, Bornemann-cimenti H, Wejbora M, Kern-Pirsch c, Michaeli K, sandner-Kiesling a. Pregaba-lin and postoperative hyperalgesia. a review. schmerz 2011;25:12-8.

30. Kim sH, stoicea N, soghomonyan s, Bergese sD. intra-operative use of remifentanil and opioid induced hyper-algesia/acute opioid tolerance: systematic review. Front Pharmacol 2014;5:108.

References

1. Mion g, Villevieille t. Ketamine pharmacology: an up-date (pharmacodynamics and molecular aspects, recent findings). CNS Neurosci Ther 2013;19:370-80.

2. elia N, tramer Mr. Ketamine and postoperative pain--a quantitative systematic review of randomised trials. Pain 2005;113:61-70.

3. Bell rF, Dahl JB, Moore ra, Kalso e. Perioperative ket-amine for acute postoperative pain. cochrane Database syst rev 2006;2006:cD004603.

4. subramaniam K, subramaniam B, steinbrook ra. Ketamine as adjuvant analgesic to opioids: a quantita-tive and qualitative systematic review. anesth analg 2004;99:482-95.

5. laskowski K, stirling a, McKay WP, lim HJ. a system-atic review of intravenous ketamine for postoperative an-algesia. can J anaesth 2011;58:911-23.

6. Jouguelet-lacoste J, la colla l, schilling D, chelly Je. the use of intravenous infusion or single dose of low-dose ketamine for postoperative analgesia: a review of the current literature. Pain Med 2015;16:383-403.

7. Martinez V, Derivaux B, Beloeil H, regional a, the Pain committee of the French society of a, intensive c. Keta-mine for pain management in France, an observational survey. anaesth crit care Pain Med 2015;34:357-61.

8. De Kock M, lavand’homme P, Waterloos H. ‘Balanced analgesia’ in the perioperative period: is there a place for ketamine? Pain 2001;92:373-80.

9. Bornemann-cimenti H, lederer aJ, Wejbora M, Michae-li K, Kern-Pirsch c, archan s, et al. Preoperative prega-balin administration significantly reduces postoperative opioid consumption and mechanical hyperalgesia after transperitoneal nephrectomy. Br J anaesth 2012;108:845-9.

10. Bergeron N, Dubois MJ, Dumont M, Dial s, skrobik Y. in-tensive care Delirium screening checklist: evaluation of a new screening tool. intensive care Med 2001;27:859-64.

11. radtke FM, Franck M, oppermann s, lutz a, seeling M, Heymann a, et al. [the intensive care Delirium screen-ing checklist (icDsc)--translation and validation of in-tensive care delirium checklist in accordance with guide-lines]. anasthesiol intensivmed Notfallmed schmerzther 2009;44:80-6.

12. schulz KF, altman Dg, Moher D, group c. coNsort 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.

13. sakai t, tomiyasu s, ono t, Yamada H, sumikawa K. Multiple sclerosis with severe pain and allodynia allevi-ated by oral ketamine. clin J Pain 2004;20:375-6.

14. Kannan tr, saxena a, Bhatnagar s, Barry a. oral keta-mine as an adjuvant to oral morphine for neuropathic pain in cancer patients. J Pain symptom Manage 2002;23:60-5.

15. Marchetti F, coutaux a, Bellanger a, Magneux c, Bour-geois P, Mion G. Efficacy and safety of oral ketamine

Funding.—this project was supported by institutional funding.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Acknowledgments.—the authors wish to thank Dr. gabriele Müller, Department of anesthesiology and intensive care Medicine, University clinic graz, for her extensive help in conducting this study.Article first published online: June 21, 2016. - Manuscript accepted: June 17, 2016. - Manuscript revised: June 14, 2016. - Manuscript received: November 2, 2015.


blood products (packed red blood cells [PrBc], fresh frozen plasma [FFP], platelets) remains a preferred treatment: it is estimated that more

Perioperative bleeding is common in cardiac surgery and is associated with increases in

morbidity and mortality.1, 2 the transfusion of


Point-of-care-based protocol with first-line therapy with coagulation factor concentrates is associated with decrease allogenic blood

transfusion and costs in cardiovascular surgery: an italian single-center experience

Davide treVisaN *, laura ZaVatti, Davide gaBBieri, Marco PeDUlli, gianbeppe giorDaNo, Marco Meli

anestesia e rianimazione, Hesperia Hospital, Modena, italy*corresponding author: Davide trevisan, Hesperia Hospital s.p.a., servizio di anestesia e rianimazione, Via arquà 80/a 41123 Modena, italy. e-mail: [email protected]


lavoro: 10955-Mastitolo breve: PoiNt-oF-care BaseD Protocol iN carDiac sUrgerYprimo autore: treVisaNpagine: 1077-88citazione: Minerva anestesiol 2016;82:1077-88

a B s t r a c tBacKgroUND: transfusion in patients having cardiac surgery has been associated with increased morbidity, mortal-ity, and costs. this analysis assessed the impact of a rotational thromboelastometry (roteM®)- and functional platelet assessment (Multiplate®)-based protocol for bleeding management on perioperative outcomes and costs in patients un-dergoing cardiac surgery.MetHoDs: this retrospective analysis of the records of all patients who underwent cardiac surgery at the Hesperia Hos-pital, Modena, italy, from December 2012 to December 2013 compared outcomes and costs of bleeding management for the two 6-month periods before/after introduction of the roteM- and Multiplate-based protocol. Descriptive and cor-relation analysis were performed as appropriate. Propensity score matching and its correlation analysis were performed.resUlts: Data from 768 consecutive patients (mean age ~69 years, ~66% male) were included; 50.7% and 49.3% of patients had surgery before and after protocol introduction, respectively. Significantly fewer patients required transfu-sions of packed red blood cells after the protocol introduction over the 24 hours postsurgery (100 vs. 197 patients; P<0.001) and during icU stay (134 vs. 221 patients; P<0.001). A significantly greater proportion of patients treated after protocol introduction received prothrombin complex concentrate (31 vs. 16; P<0.05) and fibrinogen concentrate (36 vs. 13; P<0.001). A significantly greater proportion of patients treated after protocol introduction had an ICU stay duration <48 hours (81.5% vs. 71.5%; P<0.001). roteM-based bleeding management was associated with a saving of €128,676.23 for the 379 patients undergoing surgery post-protocol introduction (€339.52 per patient).coNclUsioNs: roteM-guided bleeding management in patients undergoing cardiac surgery was cost-effective and associated with an increase of administration of coagulation factor concentrates and a decrease of icU length of stay.(Cite this article as: trevisan D, Zavatti l, gabbieri D, Pedulli M, giordano g, Meli M. Point-of-care-based protocol with first-line therapy with coagulation factor concentrates is associated with decrease allogenic blood transfusion and costs in cardiovascular surgery: an italian single-center experience. Minerva anestesiol 2016;82:1077-88)Key words: thromboelastography - Blood transfusion - cardiac surgical procedures.


treVisaN PoiNt-oF-care BaseD Protocol iN carDiac sUrgerY


in order to improve our standard of care, in 2013 we developed and instituted a new pro-tocol for hemostatic management during car-diac surgery at the Hesperia Hospital, utilizing coagulation factor concentrates as first choice therapy when suggested by roteM results. the algorithm was developed based on similar protocols and studies in the literature.11-18 to assess the impact of the new strategy on peri-operative outcomes and costs, we conducted a retrospective analysis comparing the data of patients who had cardiac surgery before the institution of the algorithm versus those who had cardiac surgery with roteM-guided he-mostatic management according to the new protocol.

Moreover, prevention of bleeding was con-sidered in the new protocol by screening pa-tients at risk of acquired thrombocytopathy (Figure 1) with preoperative platelet function assessment. Bedside platelet function testing was suggested as an option to guide treat-ment interruption rather than arbitrary use of a specified period of delay. Platelet inhibitory response to clopidogrel determines caBg-related bleeding, and a strategy based on pre-operative platelet function testing to determine the timing of caBg in clopidogrel-treated pa-tients led to ≈50% shorter waiting time than recommended in the current guidelines. 19 For these reasons, the 2012 Update of the society of thoracic surgeons guideline suggested that a delay of even a day or two is reason-able to decrease bleeding and thrombotic risk in acs patients.20 Bedside platelet function testing has been evaluated during clopidogrel exposure but might also be useful in prasug-rel- or ticagrelor-treated patients, as recently shown for prasugrel.21 at our institution, plate-let function assessment has been done with a Multiplate device (Verum Diagnostica gmbH, Munich, germany).


Study design and patients

the study design was a retrospective analy-sis of the records of all patients who underwent

than 50% of cardiac surgery patients receive a blood transfusion.3, 4 However, transfusion in patients having cardiac surgery has been as-sociated with increased morbidity (increased incidence of acute renal failure, thrombotic/thromboembolic events, transfusion-related immunomodulation with subsequent nosoco-mial infections and sepsis, transfusion-related acute lung injury, and transfusion-related cir-culatory overload), mortality, and costs.1, 5-9

several interventions have been proposed over the years to diagnose and manage hem-orrhage during cardiac surgery. Point-of-care (Poc) tests, including rotational thromboelas-tometry (roteM), have emerged as a feasible and valid approach for the management of bleeding.1, 10 a number of roteM-based al-gorithms for hemostatic management during surgery have been developed.1 evidence sug-gests that the use of these algorithms reduces the consumption of blood products and has a positive impact on costs.11-15 However, stud-ies on roteM effectiveness are limited and several aspects related to this methodology are not fully understood. the use of roteM in italy is not widespread; it is mostly used in university hospitals, in liver transplant, trauma surgery and heart surgery units. in our center, roteM was used for 8 years without an es-tablished protocol for bleeding management. subsequently, when coagulation factor con-centrates (fibrinogen, prothrombin complex) became available, they were used without a protocol and not as first choice.

as our institution refers to the transfusional service of a nearby University Hospital, we have to consider a 30-minute transportation time to add to the blood product preparation. For this reason, before introduction of the al-gorithm, coagulation factor concentrates (Hae-mocomplettan P, CLS Behring; Confidex 500, cls Behring; Uman complex 500, Kedrion) were considered as buffering therapy in emer-gent bleeding waiting for the blood product. Blood products were ordered in advance in every intervention at risk of severe acquired coagulopathy (long cPB, clinical condition). cryoprecipitate is not available according to transfusional service policy.

PoiNt-oF-care BaseD Protocol iN carDiac sUrgerY treVisaN


Figure 1.—ROTEM-guided protocol for hemostatic management during cardiac surgery. A10: amplitude of clot firmness 10 minutes after clotting time; aDP: adenosine diphosphate aggregometry test; cPB: cardiopulmonary bypass; crF: chronic renal failure; ct: clotting time; DDaVP: desmopressin; eX: eXteM test; FiB: FiBteM test; HeP: HePteM test; Ht: he-matocrit; HMs: hemostasis management system; icU: intensive care unit; iN: iNteM test; Plt: platelets (units); roteM: rotational thromboelastometry; tc: core temperature; traP: thrombin receptor-activated peptide aggregometry test.

BEFORE SURGERY

AGGREGOMETRY

DURING ANDAFTER SURGERY:

ROTEM

SURGICAL REVISION



tion.1, 22 By utilizing a variety of activators, this method provides a detailed analysis of specific aspects of the coagulation cascade and can identify specific cascade abnormalities (unlike conventional coagulation tests).23, 24 By providing graphic and numeric data on clot formation, ROTEM can detect isolated defi-cits of fibrinogen (FIBTEM test) and gen-eral coagulation factors deficiencies (INTEM and eXteM tests). Matching these results with special tests (aPteM and HePteM), roteM analysis provides information on hy-perfibrinolisys and the presence of heparin in patient blood.25, 26

a Multiplate device was used to screen platelet function in selected patients (Figure 1). By adding different activators to a pa-tient’s whole blood sample, this device pro-vides information on the grade of inhibition of several pathways of platelet aggregation ac-tivation. the asPi-test examines the arachi-donic acid pathway and is therefore sensitive to aspirin administration; the aDP-test ex-amines the aDP-receptor pathway and detect thienopyridines platelet inhibition; the traP-test examines the thrombin-receptor pathway and is affected both by gPiib/iiia inhibitors and congenital or acquired thrombocytopa-thy (e.g. uremia). Based on the experience of ranucci et al., we decided to administer des-mopressin (DDaVP) in a dose of 0.3 mg/kg in 20 minutes at skin incision in patients with an aDP-test between 30 and 50, repeatable at cPB discontinuation.21, 27 in patients with a history of recent thienopyridines administra-tion and aDP<30 we recommended to post-pone the intervention and repeat the platelet aggregometry test 24 hours later, if no urgent surgical indications were present. in every pa-tient with a traP-test <50, indicating poor platelets viability, we recommended the ad-ministration of DDaVP and platelets transfu-sion.

Bleeding management

Before introduction of the algorithm, bleed-ing was managed by administering coagulation factor concentrates and blood products based

cardiac surgery at the study center (Hesperia Hospital, Modena, italy) from December 2012 to December 2013. in May 2013, an algorithm was developed (Figure 1) for the prevention and management of bleeding during cardiac surgery undertaken at the study center. the analysis compared outcomes (transfusion re-quirement and length of intensive care unit [icU] stay) and costs of bleeding management before (from December 2012 to May 2013, 5 months) and after (from June 2013 to Decem-ber 2013, 6 months) the introduction of the al-gorithm. the study was approved by the insti-tutional ethics committee (registration number 29/16).

Anesthesia, surgery, and postoperative care

all surgical interventions, between De-cember 2012 and December 2013 were per-formed by the same team, which consisted of four surgeons and five assistants; the in-dividuals in the team did not change during the above-mentioned period. Premedication of patients was performed with midazolam 0.005-0.01 mg/kg and morphine 0.005-0.01 mg/kg. Fentanyl 1-4 µg/kg and propofol 1-3 mg/kg were administered as induction anes-thesia. Maintenance was performed with fen-tanyl 3-5 µg/kg·h and propofol 3-5 mg/kg·h tailored to clinical response and Bispectral in-dex. tranexamic acid 15 mg/kg was adminis-tered at the time of induction and after discon-tinuation of cardiopulmonary bypass (cPB). Heparin and protamine dose were adminis-tered according to the HMs Plus hemostasis management system (Medtronic, Minneapo-lis, Us) activated clotting time (act) test. recombinant antithrombin was administered preoperatively to those patients with activity <70%.

Point-of-care tests

at our institution, we routinely use roteM and Multiplate for perioperative bleeding man-agement.

roteM analyzes viscoelastic changes oc-curring in blood samples during clot forma-



Cost analysis

cost analysis considered the resources used for bleeding management, namely: blood products (PrBc, FFP, platelets); coagula-tion factors (Uman complex [prothrombin complex concentrate], Confidex [human pro-thrombin complex concentrate], Haemocom-plettan [fibrinogen], Emosint [desmopressin]); icU admission; re-operation; materials for performing the roteM assays. these re-sources were valued in 2014 euros and costs were estimated according to records kept in the center’s purchasing department, where all spending related to the cardiac surgeries, in-cluding the purchase of blood products from an outside transfusion service, was document-ed. Mean total direct costs per patient before and after introduction of the roteM-guided algorithm for hemostatic management were compared.


all statistical analyses were performed us-ing the sPss statistical package (version 20.0) and r software version 2.12. the r package of Matchit was used for the propensity score (Ps) analysis.

Data before and after introduction of the roteM-guided algorithm were analyzed de-scriptively (percentages or mean±standard de-viation [sD] for normal distributed variables; median±1st and 3rd quartile for non-normal dis-tributed variables) and using the chi-square test, analysis of variance (aNoVa) and Kruskall-Wallis Test, as appropriate. Significance was assumed if the P value was <0.05. Ps match-ing, binary logistic regression models, and Ka-plan Meier analyses were also performed. Pro-pensity matched statistics have been proposed to correlate similar samples of patients before and after the introduction of protocols. We performed caliper matching on the Ps (near-est available matching). Pairs (normalized and non-normalized groups) on the Ps logit were matched to within a range of 0.2 multiplied by the standard deviation. covariates included in the model were age, gender, weight, height,

on individual clinical judgment and experience without a protocol, in order to correct roteM analysis alteration.

after introduction of the algorithm, roteM measurements were performed at the following time points: in all high risk patients (those with liver or renal disease, or those with a body mass index >30 kg/m2) and high risk procedures (long predicted cPB times, redo operation) roteM analysis was performed before, during and after cPB; in patients who had an unexpected cPB time of >120 minutes, roteM analysis was done during and after cPB; roteM analysis was done after cPB in patients who had an absence of obvious bleed-ing causes for whom surgical hemostasis was insufficient after more than 20 minutes. In in-tensive care, severe bleeding was defined as >1.5 ml/kg/h for 3 consecutive hours on the basis that the icU team felt this was the thresh-old above which they would feel compelled to intervene.

PrBc transfusion triggers are indicated in an institutional protocol which did not change over the study period. Pre-cPB transfusion was indicated in order to achieve a predicted in-pump hematocrit (Ht) over 22%. intra-cPB transfusion were indicated in case of Ht<22%. Post-cPB and icU transfusion were manda-tory with Ht<21%, indicated with Ht<27% (based on clinical, tissue perfusion and he-modynamic evaluation), not indicated with Ht>27% in a normovolemic patient.

Variables analyzed

the following data were extracted from patient records: demographic characteristics; baseline cardiac status; preoperative use of antiplatelets and anticoagulant drugs; sur-gery characteristics; transfusion requirement (PrBc, FFP, platelets, coagulation factor concentrates) perioperatively and during the stay in the icU; length of ventilation; length of stay in the icU; need for re-operation for bleeding or tamponade; and mortality. We did not collect data on the use of vasoactive drugs and roteM results were not included in the analysis.



Results

Data from 768 consecutive patients under-going cardiac surgery were included in the study. of these, 389 (50.7%) patients had sur-gery before the institution of the new protocol for bleeding management and 379 (49.3%) pa-tients had surgery after institution of the new protocol. a total of 29 interventions were post-poned in patients showing adenosine diphos-phate (aDP) aggregometry less than 30. Pa-tient demographic and clinical characteristics are shown in table i. overall, demographic

Body Mass index (BMi), smoking status, euro-pean system for cardiac operative risk eval-uation (euroscore), canadian cardiovascu-lar society angina grading scale (ccs), New York Heart association (NYHa) Functional Classification, left ventricular ejection frac-tion (lVeF), extracorporeal circulation (ecc) time, clamp time and presence of comorbidity. Data on antiplatelet therapy were not included in the model as the use of Multiplate changed the management of these patients. the P.s. matching algorithm selected 660 matched pa-tients (330 in each group).

Table I.—�Patient demographic characteristics and cardiac status before surgery (all population and Propensity Score population).

characteristicPre-roteM-based protocol

(N.=389)

Post-roteM-based protocol

(N.=379)

Propensity score

Pre-roteM-based protocol

(N.=330)


(N.=330)

age, years 69.0±11.8 69.2±11.6 69.1±11.39 69.3±11.71gender, N. (%)

Female 131 (33.7) 134 (35.4) 110 (33.3) 115 (34.8)Male 258 (66.3) 245 (64.6) 220 (66.7) 215 (65.2)

Weight, kg 77.0±14.9 75.8±15.6 77.0±15.04 76.4±14.62Height, cm 168.0±9.6 167.3±11.7 167.9±9.5744 167.9±8.6999BMi, kg/m2 27.2±4.7 28.7±28.7 27.2±4.764 27.0±4.328euroscore 6.0±2.8 6.2±3.0 6.0±2.77 6.0±2.90ccs, N. (%)

0 238 (61.2) 213 (56.2) 197 (59.7) 182 (55.2)i 9 (2.3) 8 (2.1) 7 (2.1) 7 (2.1)ii 111 (28.5) 127 (33.5) 99 (30.0) 112 (33.9)iii 31 (8.0) 31 (8.2) 27 (8.2) 29 (8.8)

NYHa, N. (%)i 104 (26.7) 123 (32.5)* 101 (30.6) 111 (33.6)ii 193 (49.6) 138 (36.4)* 143 (43.3) 122 (37.0)iii 87 (22.4) 116 (30.6)* 83 (25.2) 95 (28.8)iV 5 (1.3) 2 (0.5)* 3 (0.9) 2 (0.6)lVeF, N. (%)

>50% 270 (69.4) 255 (67.3) 228 (69.1) 223 (67.6)30-49% 116 (29.8) 123 (32.5) 101 (30.6) 106 (32.1)<30% 3 (0.8) 1 (0.3) 1 (0.3) 1 (0.3)

comorbidities, N. (%) 382 (98.2) 355 (93.7)* 323 (97.9) 324 (98.2)Diabetes 84 (21.6) 100 (26.4) 67 (20.3)*** 89 (27.0)***Hypercholesterolemia 214 (55.0) 221 (58.3) 178 (53.9)**** 199 (60.3)****Hypertension 380 (97.7) 351 (92.6)* 321 (97.3) 320 (97.0)renal failure 2 (0.5) 10 (2.6)** 2 (0.6)***** 9 (2.7)*****cerebrovascular accidents 19 (4.9) 23 (6.1) 15 (4.5) 20 (6.1)Bacterial endocarditis 11 (2.8) 9 (2.4) 9 (2.7) 9 (2.7)cHF 2 (0.5) 0 1 (0.3) -chronic lung disease 16 (4.1) 27 (7.1) 14 (4.2) 22 (6.7)

Values are mean±standard deviation unless otherwise stated.*P<0.001 pre- vs. post-roteM protocol; **P<0.02 pre- vs. post-roteM protocol; ***P<0.05 pre- vs. post-roteM protocol; ****P<0.01 pre- vs. post-roteM protocol; *****P=0.03 pre- vs. post-roteM protocol.BMi: Body Mass index; ccs: canadian cardiovascular society angina grading scale; cHF: congestive heart failure; euroscore: euro-pean system for cardiac operative risk evaluation; lVeF: left ventricular ejection fraction; NYHa: New York Heart association functional classification; ROTEM: rotational thromboelastometry.



(56.8%) treated before protocol introduction and 182/379 (48.1%) treated after protocol introduction were transfused (P<0.01). Nine patients in each group (2.3% vs. 2.4%, respec-tively) died within 30 days of surgery.

Patients undergoing cardiac surgery after in-troduction of the new protocol for hemostatic management had a shorter duration of ventila-tion and stay in the icU than patients having surgery before the protocol introduction, and fewer patients in this group required reopera-tion compared with patients who underwent surgery before protocol institution (table iV); however, these differences between groups were not statistically significant. A significant-ly greater proportion of patients treated after protocol introduction than before had an icU stay duration shorter than 48 hours (81.5% vs. 71.5%; P<0.001) and a smaller proportion had

and baseline heart disease characteristics of the two groups were similar, although there were some differences in New York Heart associa-tion classifications and incidence of comorbid-ities. surgical characteristics of the two groups were also comparable (table ii). the majority of patients before and after protocol introduc-tion underwent elective surgery (92.5% and 92.9%) and the most frequent intervention was valve surgery (42.4% and 40.1%) followed by coronary artery bypass grafting (caBg, 24.2% and 31.4%). We excluded from analysis off-pump caBg (7 and 5 cases before and af-ter protocol introduction). antiplatelets drugs administered preoperatively are summarized in table ii.

Data related to the use of blood products and coagulation factors concentrates are sum-marized in table iii. overall, 221/389 patients

Table II.—�Surgery characteristics, preoperative coagulation testing and antiplatelet drug exposure (all population and Propensity Score population).


(N.=389)


(N.=379)

Propensity score


(N.=330)


(N.=330)

surgery characteristicsecc, minutes 99.3±41.5 98.8±50.1 98.2±41.41 97.9±48.20cross-clamp, minutes 67.5±30.2 65.5±31.0 66.1±28.39 65.6±31.24Procedure type, N. (%)

elective 360 (92.5) 352 (92.9) 306 (92.7) 310 (93.9)Urgent 4 (1.0) 3 (0.8) 4 (1.2) 2 (0.6)emergency 25 (6.4) 24 (6.3) 20 (6.1) 18 (5.5)redo surgery 15 (3.9) 26 (6.9) 12 (3.6) 22 (6.7)

surgery type, N. (%)caBg 94 (24.2) 119 (31.4) 84 (25.5) 106 (32.1)Valvular 165 (42.4) 152 (40.1) 141 (42.7) 132 (40.0)Dissection 8 (2.1) 7 (1.8) 6 (1.8) 4 (1.2)aortic arch surgery 5 (1.3) 3 (0.8) 4 (1.2) 3 (0.9)aortic root surgery 38 (9.8) 40 (10.6) 30 (9.1) 34 (10.3)caBg + valvular 49 (12.6) 37 (9.8) 40 (12.1) 34 (10.3)other 30 (7.7) 21 (5.5) 25 (7.6) 17 (5.2)

refusing transfusions, N. (%) 5 (1.3) 6 (1.6) 3 (0.9) 5 (1.5)Preoperative antiplatelet drug administration and platelet function testing

aspirin within 7 days, N. (%) 148 (38.0) 138 (36.4) 121 (36.6) 117 (35.4)clopidogrel within 5 days, N. (%) 59 (13.6) 61 (16.1) 49 (14.8) 49 (14.8)Preoperative aggregometry test, N. (%) 70 (18.0) 112 (29.6) 61 (18.5) 99 (30.0)

asP<30U 62 (15.9) 106 (28.0) 55 (16.7) 96 (29.1)aDP<30U 34 (8.7) 41 (10.8) 29 (8.8) 38 (11.5)traP<70U 26 (6.7) 38 (10.0) 22 (6.7) 34 (10.3)intervention postponed until aDP>30U none 29 (7.7) none 22 (6.7)

Values are mean±standard deviation unless otherwise stated. Urgent surgery was defined as surgery that may be delayed up to 24 hours. Emer-gency surgery was defined as surgery that should be performed as soon as possible.aDP: adenosine diphosphate aggregometry test; asP: aspirin aggregometry test; caBg: coronary artery bypass graft; ecc, extracorporeal circulation; roteM: rotational thromboelastometry; traP: thrombin-receptor activated peptide aggregometry test.



of stay in icU (65 hours vs. 61 hours, P=0.047).the results of the cost estimates of the re-

sources used for bleeding management before and after institution of the new protocol are summarized in table V. Bleeding management using the roteM-guided protocol was as-sociated with a saving of €128,676.23 for the 379 patients undergoing surgery postprotocol introduction (€339.52 per patient). this was mainly triggered by a decreased use of PrBc, shorter duration of icU stay, and reduction in the number of reoperations.

Discussion

this manuscript describes the results of a retrospective analysis aiming to determine the

an icU stay duration from 48 hours to 7 days (11.3% vs. 23.1%, respectively; P<0.001).

the mean hemoglobin level before surgery and at icU discharge was similar in the two groups (respectively 12.4±1.6 and 9.1±0.9 mg/dl preprotocol versus 12.7±1.8 and 8.9±1.0 mg/dl postprotocol introduction). stroke was diagnosed in one patient in each group. No postoperative myocardial infarction due to acute graft obstruction was registered in the two groups.

Propensity score matching results are also reported in table i, ii, iii and iV. Kaplan-Mei-er analysis on the Ps population showed a sig-nificantly higher pre-ROTEM estimated time of ventilation (31 hours vs. 28 hours, P<0.001) and a significantly lower post-ROTEM length

Table III.—�Use of blood products and coagulation factor concentrates (all population and Propensity Score popula-tion).


(N.=389)


(N.=379)

Propensity score


(N.=330)


(N.=330)

total erythrocytes UnitsPatients (N., %) 221 (56.8) 182 (48.1)* 186 (56.4) 159 (48.3)***Median (min-max) 1 (0-95) 0 (0-30) 1 (0-21) 0 (0-30)1st-3rd quartiles 0-3 0-2 0-3 0-2

24h erythrocytes UnitsPatients (N., %) 197 (50.6) 100 (26.4)** 164 (49.7) 88 (26.7)**Median (min-max) 1 (0-17) 0 (0-12)** 0 (0-15) 0 (0-10)**1st-3rd quartiles 0-2 0-1 0-2 0-1

icU erythrocytes UnitsPatients (N., %) 221 (56.8) 134 (35.4)** 186 (56.4) 119 (36.1)**Median (min-max) 1 (0-83) 0 (0-29)** 1 (0-21) 0 (0-29)**1st-3rd quartiles 0-3 0-2 0-2 0-2

Fresh Frozen Plasma UnitsPatients (N., %) 33 (8.5) 21 (5.6) 26 (7.9) 17 (5.2)Median (min-max) 0 (0-37) 0 (0-42) 0 (0-6) 0 (0-42)1st-3rd quartiles 0-0 0-0 0-0 0-0

Platelets UnitsPatients (N., %) 23 (5.9) 28 (7.4) 17 (5.2) 23 (7.0)Median (min-max) 0 (0-18) 0 (0-5) 0 (0-4) 0 (0-5)1st-3rd quartiles 0-0 0-0 0-0 0-0

Human complex unitsPatients, N. (%) 1 0 0 0Mean ± sD (range) 4.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0

Confidex unitsPatients, N. (%) 16 31*** 11 26***Mean ± sD (range) 1.9 ± 0.7 (1-3) 2.1 ± 1.1 (1-4) 1.6 ± 0.6 (1-3) 2.2 ± 1.0 (1-4)

Haemocomplettan unitsPatients, N. (%) 13 36** 9 29**Mean ± sD (range) 1.8 ± 0.6 (1-3) 2.4 ± 1.0 (1-6) 1.9 ± 0.6 (1-3) 2.6 ± 1.0 (1-6)

Values are mean±standard deviation (range) unless otherwise stated.*P<0.02 pre- vs. post-roteM protocol; **P<0.001 pre- vs. post-roteM protocol; ***P<0.05 pre- vs. post-roteM protocol.icU: intensive care unit; roteM: rotational thromboelastometry.



costs. the analysis showed that introduction of the protocol led to a reduction in the number of units of PrBc administered over 24 hours and

impact of a roteM-based protocol for the management and prevention of surgery-asso-ciated bleeding on perioperative outcomes and

Table IV.—�Duration of mechanical ventilation, length of stay in the intensive care unit, and need for revision surgery (all population and Propensity Score population).


(N.=389)


(N.=379)

Propensity score


(N.=330)


(N.=330)

Duration of ventilation, hoursMedian (min-max)1st-3rd quartiles

10 (2-2520)(6-10)

8 (2-2880)*(7-10)

10 (2-2880)*(7-10)

8 (2-1430)(6-10)

length of icU stay, hoursMedian (min-max)1st-3rd quartiles

19 (3-4680)(18-66)

18 (2-2400)**(16-40)

20 (3-1472)(18-66)

18 (3-2400)***(16-40)

length of icU stay<48 hours, N. (%)48 hours – 7 days, N. (%)>7 days, N. (%)

278 (71.5)90 (23.1)21 (5.4)

309 (81.5)*43 (11.3)*27 (7.1)

239 (72.4)75 (22.7)16 (4.8)

266 (80.6)**41 (12.4)23 (7.0)

Patients needing revision surgery, N. (%) 20 (5.1) 11 (2.9) 13 (3.9) 10 (3.0)*P<0.001 **P=0.002 ***P<0.05 vs. post-roteM-based protocol.icU: intensive care unit; roteM: rotational thromboelastometry.

Table V.—�Cost estimates before and after introduction of a ROTEM-guided protocol.

treatment Unit cost (€)Units pre-

roteM-based protocol (N.)

cost pre-roteM-based

protocol (€)

Units post-roteM-based protocol (N.)

cost post-roteM-based

protocol (€)

PBrc 188.65 891 168087.15 701 132243.65FFP 161.00 132 21252.00 121 19481.00Platelets 203.00 54 10962.00 45 9135.00icU staya 104.16b 9.96 x 389 403662.97 7.69 x 379 303575.36re-intervention for bleeding 2775.00c 20 55500.00 11 30525.00coagulation factor concentrate

UMaN coMPleX 120.00 4 480.00 0 0.00coNFiDeX 139.00 30 4170.00 65 9035.00HaeMocoMPlettaN 400.00 23 9200.00 85 34,000.00

other protocol drugseMosiNt 1.83 430 786.90 622 1138.26

consumable materials for roteMcuvette 1483.20 5 7416.00 3 4449.60eXteM 432.00 5 2160.00 2 864.00eXteM s 193.92 0 0.00 14 2714.88FiBteM 320.00 7 2240.00 2 640.00FiBteM s 304.80 0 0.00 14 4267.20HePteM 480.00 2 960.00 0 0.00HePteM s 319.20 0 0.00 11 3511.20iNteM 432.00 6 2592.00 0 0.00iNteM s 193.92 0 0.00 14 2714.88starteM 162.00 7 1134.00 3 486.00

total consumable costs 16,502.00 19,647.76total cost 707,105.02 578,428.79total cost DiFFereNce -128,676.23cost DiFFereNce Per PatieNt -339.52acalculated based on an 8-hour nurse shift multiplied by the number of patients, and divided by 2 (2 beds covered per nurse).bPer 8-hour nurse shiftcPer interventionFFP: fresh frozen plasma; icU: intensive care unit; PrBc: packed red blood cells; roteM: rotational thromboelastometry.



literature. in one turkish study, patients un-dergoing elective caBg either had clinician-directed transfusion, or transfusion guided by a routine thromboelastography algorithm; use of the algorithm resulted in a reduction in blood products (FFP, platelets) versus clinician-directed management.31 another group of patients undergoing cPB demon-strated reduced transfusion requirement and 24-hour postoperative bleeding with the use of a thromboelastometry test-guided algo-rithm.14 several studies investigated the use of coagulation management algorithms in set-tings including aortic arch replacement/aortic dissection, cardiovascular surgery, trauma and transplant surgery; 12, 13, 32 these studies also found that use of an algorithm reduced transfusion requirements, thromboembolic/thrombotic events, and the incidence of mas-sive transfusion. a recently published review article,33 reported on 16 studies (including 8507 cardiac surgical patients overall) deal-ing with transfusion protocols in cardiovascu-lar surgery comparing Poc-based algorithms versus algorithms based on routine labora-tory testing and clinician discretion or stan-dard of care. all 16 studies demonstrated a reduction in transfusion requirements in the Poc group. the effect size was dependent on study population (simple caBg surgery or complex cardiac/aortic surgery), the aver-age blood loss, the extent of Poc diagnos-tics (viscoelastic tests only or viscoelastic tests plus Multiplate), and the first-line use of coagulation factor concentrates. seven stud-ies reported a reduction in the incidence of mediastinal re-exploration and a reduction in hospital costs.12, 32, 34-36 our observations are in agreement with these results: the signal of reduction in allogenic blood transfusion and costs provided by the adoption of a Poc-based protocol seems to be amplified by the first-line therapy with coagulation factor con-centrates.

the present study also demonstrated a re-duction in costs, but it should be noted that the actual cost reduction is most likely less than reported here. the calculation of costs assumed that bed saturation in the icU was

during the ICU and hospital stay, and a signifi-cant increase in the number of patients leaving the icU within 48 hours postsurgery. Use of the protocol led to a decrease in costs of ap-proximately €339 per patient.

When comparing the number of units of FFP and platelets transfused, there was no sta-tistically significant difference before and after the roteM protocol introduction. that the change in these endpoints is not more marked is likely due to the fact that a roteM-based strategy utilizing aggregometry and coagula-tion factor concentrates was already in place at the center, albeit one that was used in a far less organized manner. Moreover, the number of interventions postponed because of aDP aggregometry less than 30 (29 patients) could be the reason for the slight increase of platelets transfused. the introduction of a more regu-lated protocol resulted in the increased use of coagulation factor concentrates rather than a change in use of FFP and platelets.

Notably, better coagulation control post-protocol resulted in a reduction in the number of patients transfused during the periopera-tive 24 hours, icU stay and hospital stay, and a greater proportion of patients staying in the icU for less than 48 hours. transfusions are as-sociated with several complications, including transfusion reactions, acute lung injury, circu-latory overload, and immunomodulation;28 the decrease in the number of per-patient transfu-sions may be the first step in reducing transfu-sion-related complications.

the protocol implemented was developed based on the clinical experience in the center and published literature on the topic of man-aging surgical bleeding.1, 12, 13, 15, 27, 29, 30 since roteM and Multiplate aggregometry were established methods at our center, and coagu-lation factor concentrates were already in use, implementation of the protocol was relatively simple. the use of the protocol has increased the use of coagulation factor concentrates, and given the center’s staff common objectives, re-sulting in a greater uniformity of behavior and an improvement in post-surgical outcomes.

the improvements in outcomes seen in this study are similar to those seen in the



Key messages

— a Poc-guided treatment of periop-erative bleeding in cardiac surgery posi-tively affected time of mechanical ventila-tion and icU stay.

— a Poc-based protocol was associ-ated with a significant reduction of PBRC transfused perioperatively.

— administration of coagulation fac-tor concentrates according to a Poc-based protocol has proven to be cost-effective.

References

1. görlinger K, shore-lesserson l, Dirkmann D, Hanke aa, rahe-Meyer N, tanaka Ka. Management of hem-orrhage in cardiothoracic surgery. J cardiothorac Vasc anesth 2013;27:s20-34.

2. Murphy gJ, Pike K, rogers ca, Wordsworth s, stokes ea, angelini gD, et al. liberal or restrictive transfusion after cardiac surgery. N engl J Med 2015;372:997-1008.

3. Bennett-guerrero e, Zhao Y, o’Brien sM, Ferguson tB, Jr., Peterson eD, gammie Js, et al. Variation in use of blood transfusion in coronary artery bypass graft surgery. JaMa 2010;304:1568-75.

4. Murphy MF, Murphy gJ, gill r, Herbertson M, allard s, grant-casey J. National comparative audit of blood transfusion: 2011 audit of blood transfusion in adult car-diac surgery. Birmingham, United Kingdom: National Health service, 2013 [internet]. available from: http://hospital.blood.co.uk/ [cited 2015, aug 13].

5. christensen Mc, Krapf s, Kempel a, von Heymann c. costs of excessive postoperative hemorrhage in cardiac surgery. J thorac cardiovasc surg 2009;138:687-93.

6. Murad MH, stubbs Jr, gandhi MJ, Wang at, Paul a, er-win PJ, et al. the effect of plasma transfusion on morbid-ity and mortality: a systematic review and meta-analysis. transfusion (Paris) 2010;50:1370-83.

7. Murphy gJ, reeves Bc, rogers ca, rizvi si, culliford l, angelini gD. increased mortality, postoperative mor-bidity, and cost after red blood cell transfusion in patients having cardiac surgery. circulation 2007;116:2544-52.

8. stainsby D, Jones H, asher D, atterbury c, Boncinelli a, Brant l, et al. serious hazards of transfusion: a dec-ade of hemovigilance in the UK. transfus Med rev 2006;20:273-82.

9. stanworth sJ, Brunskill sJ, Hyde cJ, Mcclelland DB, Murphy MF. is fresh frozen plasma clinically effective? a systematic review of randomized controlled trials. Br J Haematol 2004;126:139-52.

10. Davidson sJ, Mcgrowder D, roughton M, Kelleher aa. can roteM thromboelastometry predict postoperative bleeding after cardiac surgery? J cardiothorac Vasc an-esth 2008;22:655-61.

11. rahe-Meyer N, solomon c, Hanke a, schmidt Ds, Kno-erzer D, Hochleitner g, et al. Effects of fibrinogen con-centrate as first-line therapy during major aortic replace-ment surgery: a randomized, placebo-controlled trial. anesthesiology 2013;118:40-50.

12. görlinger K, Dirkmann D, Hanke aa, Kamler M, Ko-ttenberg e, thielmann M, et al. First-line therapy with

always 100%, while data from the center shows that actual total bed saturation is 65-70%; this means that the cost savings are likely to be proportionally less.37, 38 However, studies have reported that excessive post-op-erative bleeding is associated with significant costs, and any intervention resulting in lower bleeding rates is likely to lead to substantial cost savings.5 in addition, cost reduction has been reported in studies similar to the present analysis,32, 34 with savings of €2757 per case in one study of a roteM-based algorithm,32 and an overall combined saving of 44% in an-other.34 overall, these results suggest that the use of a roteM-based algorithm for bleeding management is, nevertheless, still likely to be cost saving.


this study has several limitations associated with the retrospective nature of the design, the short follow-up period, and that it took place at a single center.

Conclusions

in conclusion, the introduction of a Poc-based protocol with first-line therapy with coagulation factor concentrates for the man-agement of bleeding during cardiac surgery in a single center in italy was associated with a significant decrease in the use of PRBC and shorter icU stay.

the protocol also led to a decrease in costs associated with bleeding management during surgery, suggesting that it is an effective and potentially cost-saving option for the manage-ment of bleeding in this setting.

taking into consideration the increasing number of studies reporting positive clinical and economical results linked to Poc-based management of bleeding with coagulation factor concentrates in cardiovascular surgery patients, it is our opinion that this strategy should be considered as a standard of care in this setting. However, these results should be confirmed by a multicenter prospective ran-domized trial.



ce, Kozar ra, et al. admission rapid thrombelastog-raphy can replace conventional coagulation tests in the emergency department: experience with 1974 consecu-tive trauma patients. ann surg 2012;256:476-86.

25. Dirkmann D, gorlinger K, Peters J. assessment of early thromboelastometric variables from extrinsically activat-ed assays with and without aprotinin for rapid detection of fibrinolysis. Anesth Analg 2014;119:533-42.

26. Mittermayr M, Velik-salchner c, stalzer B, Margreiter J, Klingler a, streif W, et al. Detection of protamine and heparin after termination of cardiopulmonary bypass by thrombelastometry (roteM): results of a pilot study. anesth analg 2009;108:743-50.

27. Di Dedda U, ranucci M, Baryshnikova e, castelvec-chio s. thienopyridines resistance and recovery of platelet function after discontinuation of thienopyrid-ines in cardiac surgery patients. eur J cardiothorac surg 2014;45:165-70.

28. Katz ea. Blood transfusion: friend or foe. aacN adv crit care 2009;20:155-63.

29. crescenzi g, landoni g, Biondi-Zoccai g, Pappalardo F, Nuzzi M, Bignami e, et al. Desmopressin reduces trans-fusion needs after surgery: a meta-analysis of randomized clinical trials. anesthesiology 2008;109:1063-76.

30. ranucci M, Baryshnikova e, colella D. Monitoring pro-hemostatic treatment in bleeding patients. semin thromb Hemost 2012;38:282-91.

31. ak K, isbir cs, tetik s, atalan N, tekeli a, aljodi M, et al. thromboelastography-based transfusion algorithm reduces blood product use after elective caBg: a pro-spective randomized study. J card surg 2009;24:404-10.

32. Hanke aa, Herold U, Dirkmann D, tsagakis K, Jakob H, görlinger K. thromboelastometry based early goal-directed coagulation management reduces blood transfu-sion requirements, adverse events, and costs in acute type a aortic dissection: a pilot study. transfus Med Hemother 2012;39:121-8.

33. görlinger K, Dirkmann D, Hanke aa. Potential value of transfusion protocols in cardiac surgery. curr opin an-aesthesiol 2013;26:230-43.

34. spalding gJ, Hartrumpf M, sierig t, oesberg N, Kir-schke cg, albes JM. cost reduction of perioperative co-agulation management in cardiac surgery: value of “bed-side” thrombelastography (roteM). eur J cardiothorac surg 2007;31:1052-7.

35. Weber cF, gorlinger K, Meininger D, Herrmann e, Bin-gold t, Moritz a, et al. Point-of-care testing: a prospective, randomized clinical trial of efficacy in coagulopathic car-diac surgery patients. anesthesiology 2012;117:531-47.

36. spiess BD, gillies Bs, chandler W, Verrier e. changes in transfusion therapy and reexploration rate after institu-tion of a blood management program in cardiac surgical patients. J cardiothorac Vasc anesth 1995;9:168-73.

37. tierney lt, conroy KM. optimal occupancy in the icU: a literature review. aust crit care 2014;27:77-84.

38. Wunsch H, Wagner J, Herlim M, chong DH, Kramer aa, Halpern sD. icU occupancy and mechanical ventilator use in the United states. crit care Med 2013;41:2712-9.

coagulation factor concentrates combined with point-of-care coagulation testing is associated with decreased al-logeneic blood transfusion in cardiovascular surgery: a retrospective, single-center cohort study. anesthesiology 2011;115:1179-91.

13. görlinger K, Fries D, Dirkmann D, Weber cF, Hanke aa, schöchl H. reduction of fresh frozen plasma re-quirements by perioperative point-of-care coagulation management with early calculated goal-directed therapy. transfus Med Hemother 2012;39:104-13.

14. rahe-Meyer N, Pichlmaier M, Haverich a, solomon c, Winterhalter M, Piepenbrock s, et al. Bleeding manage-ment with fibrinogen concentrate targeting a high-nor-mal plasma fibrinogen level: a pilot study. Br J Anaesth 2009;102:785-92.

15. görlinger K, Jambor c, Hanke aa, Dirkmann D, ad-amzik M, Hartmann M, et al. Perioperative coagulation management and control of platelet transfusion by point-of-care platelet function analysis. transfus Med Hemoth-er 2007;34:396-411.

16. Wademan BH, galvin sD. Desmopressin for reducing postoperative blood loss and transfusion requirements following cardiac surgery in adults. interact cardiovasc thorac surg 2014;18:360-70.

17. Kremke M, tang M, Bak M, Kristensen Kl, Hindsholm K, andreasen JJ, et al. antiplatelet therapy at the time of coronary artery bypass grafting: a multicentre cohort study. eur J cardiothorac surg 2013;44:e133-40.

18. sousa-Uva M, storey r, Huber K, Falk V, leite-Moreira aF, amour J, et al. expert position paper on the man-agement of antiplatelet therapy in patients undergo-ing coronary artery bypass graft surgery. eur Heart J 2014;35:1510-4.

19. Mahla e, suarez ta, Bliden KP, rehak P, Metzler H, sequeira aJ, et al. Platelet function measurement-based strategy to reduce bleeding and waiting time in clopidog-rel-treated patients undergoing coronary artery bypass graft surgery: the timing based on platelet function strat-egy to reduce clopidogrel-associated bleeding related to caBg (target-caBg) study. circ cardiovasc interv 2012;5:261-9.

20. Ferraris Va, saha sP, oestreich JH, song HK, rosengart t, reece tB, et al. 2012 update to the society of tho-racic surgeons guideline on use of antiplatelet drugs in patients having cardiac and noncardiac operations. ann thorac surg 2012;94:1761-81.

21. ranucci M, Baryshnikova e, soro g, Ballotta a, De Benedetti D, conti D, et al. Multiple electrode whole-blood aggregometry and bleeding in cardiac surgery patients receiving thienopyridines. ann thorac surg 2011;91:123-9.

22. luddington rJ. thrombelastography/thromboelastom-etry. clin lab Haematol 2005;27:81-90.

23. Hincker a, Feit J, sladen rN, Wagener g. rotational thromboelastometry predicts thromboembolic com-plications after major non-cardiac surgery. crit care 2014;18:549.

24. Holcomb JB, Minei KM, scerbo Ml, radwan Za, Wade

Funding.—Medical writing assistance was funded by csl Behring, italy.Conflicts of interest.—cls Behring italy, Medtronic italy, sorin italy provided funding for medical congress fees and facilities to Davide trevisan, laura Zavatti, gianbeppe giordano and Marco Meli. Medtronic italy and sorin italy provided funding for medi-cal congress and facilities to D. gabbieri and M. Pedulli cls Behring, Janssen italy, sorin italy and Btc Medical europe provided funding to Hesperia Hospital for the organization of the medical congress “PBM nel paziente chirurgico” (28th November 2015).Acknowledgements.—We wish to thank sheridan Henness of springer Healthcare communications, auckland, New Zealand, who provided medical writing assistance on behalf of csl Behring, italy, and ray Hill, an independent medical writer, who provided journal styling prior to submission, on behalf of csl Behring, italy.Article first published online: May 18, 2016. - Manuscript accepted: May 17, 2016. - Manuscript revised: May 13, 2016. - Manuscript received: september 25, 2015.


in clinical practice and is highly effective in preventing pain at the site of tissue injury.

Failure of las to produce blockade of noci-ception is uncommon and usually due to medi-cation not being delivered to the targeted nerve in sufficient quantity, dispersion of LA from the site of action, an acidic environment limit-ing the concentration of the freebase form, or inability of the la to access its binding site on the cytoplasmic pore of voltage-gated sodium

local anesthetics (las) are a class of medi-cations that render tissue insensitive to

pain by interrupting transmission of stimuli from peripheral nerves to the brain.1 las work by blocking sodium conductance through volt-age-gated sodium channels, thereby preventing an action potential from relaying information to the brain.2, 3 la administration is common


Whole-exome sequencing of a family with local anesthetic resistance

Nathan cleNDeNeN 1, Ashley D. CANNON 2, Steven PORTER 3, christopher B. roBarDs 3, Alexander S. PARKER 4, Steven R.CLENDENEN 3 *

1Yale University School of Medicine, New Haven, CT, USA; 2Department of Genetics at the University of Alabama, Birmingham, AL, USA; 3Department of Anesthesiology, Mayo Clinic, Jacksonville, FL, USA; 4Department of Health Sciences Research, Mayo Clinic, Jacksonville, FL, USA*Corresponding author: Steven R. Clendenen, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA. E-mail: [email protected]


Lavoro: 11279-MAStitolo breve: EXOME OF A FAMILY WITH LOCAL ANESTHETIC RESISTANCEprimo autore: CLENDENENpagine: 1089-97citazione: Minerva Anestesiol 2016;82:-97

a B s t r a c tBACKGROUND: Local anesthetics (LA) work by blocking sodium conductance through voltage-gated sodium chan-nels. Complete local anesthetic resistance is infrequent, and the cause is unknown. Genetic variation in sodium channels is a potential mechanism for local anesthetic resistance. A patient with a history of inadequate loss of sensation following la administration underwent an ultrasound-guided brachial plexus nerve block with a complete failure of the block. We hypothesized that la resistance is due to a variant form of voltage-gated sodium channel.METHODS: Whole-Exome Sequencing. The patient and her immediate family provided consent for exome sequencing, and they were screened with a questionnaire to identify family members with a history of LA resistance. Exome sequenc-ing results for four individuals were referenced to the 1000 Genomes Project and the NHLBI ESP to identify variants associated with local anesthetic resistance present in less than 1% of the general population and located in functional regions of the genome.RESULTS: Exome sequencing of the four family members identified one genetic variant in the voltage-gated sodium channel shared by the three individuals with LA resistance but not present in the unaffected family member. Specifically, we noted the A572D mutation in the SCN5A gene encoding for Nav1.5.CONCLUSIONS: We identified a genetic variant that is associated with LA resistance in the gene encoding for Nav1.5. We also demonstrate that Nav1.5 is present in human peripheral nerves to support the plausibility that an abnormal form of the Nav1.5 protein could be responsible for the observed local anesthetic resistance.(Cite this article as: Clendenen N, Cannon AD, Porter S, Robards CB, Parker AS, Clendenen SR. Whole-exome sequencing of a family with local anesthetic resistance. Minerva ANestesiol 2016;82:1089-97)Key words: Exome - Anesthesia, local - Peripheral nerves - NAV1.5 Voltage-Gated Sodium Channel.

comment in p. 1029.

Minerva Anestesiologica 2016 October;82(10):1089-97© 2016 EDIZIONI MINERVA MEDICAOnline version at http://www.minervamedica.it

cleNDeNeN EXOME OF A FAMILY WITH LOCAL ANESTHETIC RESISTANCE

1090 MiNerVa aNestesiologica October 2016

on her left anterior elbow with preoperative mag-netic resonance imaging indicating a lipoma.

the anesthesia plan consisted of an ultra-sound-guided supraclavicular brachial plexus nerve block and intravenous sedation. a son-oSite M-Turbo ultrasound unit (SonoSite Inc, Bothell, WA, USA) with a high-frequency 5 MHz - 12 MHz linear transducer was used to identify the subclavian artery, first rib, and brachial plexus. a supraclavicular brachial plexus injection was performed with an in-plane lateral to medial ultrasound approach using a 22-gauge, 50 mm needle (Stimuplex R; B.Braun, Bethlehem, PA, USA). A total of 30 mL of 1.5% mepivacaine was administered, with the initial 15 mL placed in the corner bor-dered by the subclavian artery medially and the first rib inferiorly and the remainder was injected superiorly (Figure 1).

The patient was observed for 30 minutes for sensory or motor changes, and complete failure of the block was noted. the lipoma was excised without complication under general anesthesia, and at the end of the procedure the surgeon in-filtrated the wound with 12 mL of 0.5% ropi-vacaine. Ninety minutes after the block was performed, the patient had no sensory or motor block of her left arm, forearm, or hand.

Participant questionnaire

The proband’s father, mother, and maternal half-sister completed a questionnaire to gather

channels. In a prospective study of 2000 cases, failure to attain spinal anesthesia by intrathecal injection of LA occurred in approximately 3% of cases.4 the most common reason for failure was inadequate sensory level for the surgery; although, 12 cases had complete failure despite confirmation of cerebral spinal fluid flow before and after injection of la. complete resistance despite skin infiltration and confirmed drug de-livery has been reported in a small number of cases,5-7 in patients with ehlers-Danlos type III syndrome,8, 9 and in a larger screening study where patients self-reported la resistance.10 The cause of such resistance is unknown; how-ever, genetic variation is a proposed mechanism.

We hypothesized that la resistance could be due to a variant form of voltage-gated sodium channel that is incompletely inhibited by las. We tested this hypothesis by clinically confirm-ing complete la resistance in a patient present-ing for elective surgery. We surveyed the pa-tient’s immediate family for similar resistance to LAs and performed whole-exome sequencing to screen for genetic variants in voltage-gated sodium channels. We referenced their exome se-quences to publically available genetic databas-es and screened for pathogenic mutations. We identified a genetic variant in the SCN5A gene that is shared by these affected family members and may be associated with la resistance. We also performed immunohistochemistry on ca-daveric human peripheral nerves, dorsal root ganglia, spinal cord, and cerebral cortex to com-pare the distribution of Nav1.5 and Nav1.7 so-dium channels in human tissue to demonstrate that Nav1.5 is detectable on peripheral nerves, dorsal root ganglia, and cerebral cortex.


the Mayo clinic institutional review Board approved the study (IRB 12-000139) and all study participants provided written informed consent.

Identification of a patient with local anesthetic resistance

A 37-year-old woman with American Society of anesthesiologist physical status ii had a mass

Figure 1.—Ultrasound image of the patient’s brachial plexus in contact with local anesthetic, which is visible as the hyp-oechoic fluid-tissue interface surrounding the needle.

EXOME OF A FAMILY WITH LOCAL ANESTHETIC RESISTANCE cleNDeNeN


on Qubit (invitrogen) and agilent Bioanalyz-er DNa 1000 chip. the proband library was sequenced at an average coverage of ~250X and other family members were sequenced at an average coverage of 60-120X following Il-lumina’s standard protocol using the illumina cBot and cBot Paired end cluster kit version 3. The flow cells were sequenced as 101 X 2 paired end reads on an Illumina HiSeq 2000 using TruSeq SBS sequencing kit version 3 and HCS v2.0.12 data collection software. Base-calling was performed using illumina’s RTA version 1.17.21.3.

Read alignment and variant calling

The read QC and alignment, variant call-ing, and variant/gene annotation were per-formed using an in-house analytic pipeline GenomeGPS (GGPS). Specifically, read qualities were examined by FastQC.12 the 100 base paired-end reads were mapped to Human Reference Genome build 37 using Novoalign (Novocraft technologies sdn Bhd, Selangor, Malaysia). The duplicated reads were removed using the saMtools’s rmdup method.13 the alignment was then im-proved using the Genome Analysis ToolKit (GATK)14 for local realignments and quality score recalibration. single nucleotide vari-ants (sNVs) and small insertions and dele-tions (INDELs) were called using GATK’s UnifiedGenotyper. The variants were anno-tated to provide information including: 1) variant physical positions; 2) calling informa-tion (read depths, depth of alternative allele supporting reads, average mapping quality score etc.); 3) impact of variants on genes/proteins using SNPEFF 15 and the variant functional predictions using SIFT and Poly-Phen.16 all variants passed Qc were included in the genetic Variant analysis. all variants were categorized into three tiers: Tier-I vari-ants, consisting of splice/nonsense/frameshift variants, are the most likely to be pathogenic; tier-ii variants (missense or nonsynonymous variants, as well as SNVs/INDELs located in promoter and other regulatory domains) and Tier-III variants (synonymous mutations,

demographic and medical information. the demographic information included age, gen-der, height, and weight. The medical questions were designed for “yes” or “no” responses, and, if they answered “yes,” the participants were asked to explain their answer in more de-tail in a space provided. These questions were included to determine which relatives exhib-ited la resistance as well as to reveal any clinical features that may be linked with genes known to be associated with pain perception/analgesia. Specifically, the medical questions asked whether the participants:1) experienced pain after la was administered during a den-tal or other medical procedures; 2) felt that he/she is more sensitive to pain than others; 3) felt that he/she is less sensitive to pain than others; 4) felt that he/she is more sensitive to hot or cold temperatures than others; 5) felt that he/she is less sensitive to hot or cold temperatures than others; or 6) had seizures/epilepsy, schizophrenia, headaches/migraines, chronic pain, heart arrhythmia, or other neuro-logic conditions.

Whole-exome sequencing

study participants provided DNa samples via an oragene saliva kit for whole-exome sequencing. The Mayo Clinic Molecular Bi-ology core Facility performed whole-exome sequencing. Paired-end libraries were pre-pared using 1.0 µg of genomic DNa follow-ing the manufacturer’s protocol (agilent) using the Agilent Bravo liquid handler. The concentration and size distribution of the completed libraries was determined using an agilent Bioanalyzer DNa 1000 chip (santa Clara, CA, USA) and Qubit fluorometry (In-vitrogen, Carlsbad, CA, USA). Whole-exon capture was carried out using 750 ng of the prepped library following the protocol for Agilent’s SureSelect Human All Exon v4 + UTRs 71 MB kit.11 The purified capture prod-ucts were then amplified using the SureSelect Post-Capture Indexing forward and Index PCR reverse primers (Agilent) for 12 cycles. the concentration and size distribution of the completed captured libraries was determined



ing for elective surgery involving her upper extremity. the patient reported a history of failed LA nerve blocks placed by her dentist, inadequate loss of sensation with subcutane-ous LAs for intravenous access, and failed block to la placed by a cardiologist prior to cardiac catheterization. an ultrasound-guided supraclavicular brachial plexus block failed despite confirmation of LA spread (Figure 1).

a detailed evaluation of the proband’s medical history revealed a lack of clinical symptoms that are linked with genes known to be associated with pain perception/anal-gesia. the proband reported a family history of la resistance in her mother and maternal half-sister (Figure 2). these family members and her reportedly unaffected father were con-tacted and provided demographic information, medical information, and tissue samples. The mother, age 65, reported LA resistance, high pain tolerance, and sensitivity to hot and cold temperatures. She denied a history of seizures, schizophrenia, headaches, chronic pain, heart arrhythmia, and other neurologic symptoms. The proband’s maternal half-sister, age 43, also reported la resistance. she described failure of a lower extremity nerve block for ankle surgery and had received multiple la injections during dental procedures with brief and inadequate loss of sensation. She ex-pressed variable pain tolerance and increased

variants in 5’UTR and 3’UTR regions, and intergenic/intronic variants) are expected to be the most numerous.

Genetic variant analysis

Variant analysis was performed using in-genuity Variant Analysis software (Qiagen, Hilden, Germany). To identify potential caus-ative variants for LA, variants were were se-lected that segregated in affected family mem-bers. From these, variants were excluded if not located in coding or functional regions of the genome. Variants were further excluded if the frequency was greater than 1% in the public genome variant databases dbSNP and 1000 genomes.17 the remaining variants were ana-lyzed in silico for effect on protein function using the siFt 18 and PolyPhen-2 19 mutation-prediction programs.

Immunohistochemistry

Cadaveric human peripheral nerve tissue, dorsal root ganglion, spinal cord, and brain without evidence of disease was labeled with polyclonal rabbit anti-human Nav1.5 and Nav1.7 at a 1:50 and 1:400 dilution, respectively (Alo-mone laboratories). secondary staining with rabbit labelled polymer (DAKO Cytomation) was performed with 3,3’-Diaminobenzidine (DAB +, DAKO Cytomation) followed with a light gill 1 hematoxylin counterstain. antibody specificity was confirmed with a targeted block-ing peptide specific to each antigen (Alomone laboratories). Normal human myocardium was labeled as a positive control for Nav1.5 and nor-mal peripheral nerve was labeled as a positive control for Nav1.7. Immunohistochemistry on peripheral nerves was repeated in three separate individuals from different sites (left leg ampu-tation, neuroma, and left leg sural nerve).

Results

Identifying the proband and familial screening

a patient with clinically apparent la resis-tance was identified incidentally after present-

Figure 2.—Pedigree demonstrating the relationship between the subjects who had their exomes sequenced. The affected individuals reported a history of reduced efficacy of local anesthetics during medical and dental procedures. arrow in-dicates the proband.



nerves, we performed immunohistochemistry with highly specific antibodies on healthy hu-man tissue (Figure 3). Staining of peripheral nerve tissue revealed Nav1.5 immunoreactivity

doses of analgesic medications to relieve pain. she denied sensitivity to hot and cold temperatures, seizures, schizophrenia, heart arrhythmia, and other neurologic symptoms. The proband’s father, age 68, confirmed ap-propriate sensation loss with LA. He reported a subjectively high pain tolerance and denied sensitivity to hot and cold temperatures, sei-zures, schizophrenia, headaches, and other neurologic symptoms. The proband, mother, and maternal half-sister have a positive history of LA resistance, which indicated a possible genetic etiology.

Exome sequencing and genetic analysis

Whole-exome sequencing of the four fam-ily members was performed to identify ge-netic variants shared by the three individuals who had la resistance but not shared in the unaffected family member. three hundred and ninety-six genetic variants associated with 225 genes segregated in the affected relatives.

Variants located in candidate genes associ-ated with pain and analgesia are listed in table I. Only one variant was identified in a voltage-gated sodium channel susceptible to la in-hibition (Nav1.5). Specifically, we noted the A572D mutation in the SCN5A gene encoding for Nav1.5.

Nav1.5 in peripheral nerves and central ner-vous system

Nav1.5 has been extensively studied in cardiac but not nervous tissue. to deter-mine whether Nav1.5 is present in peripheral

Figure 3.—Immunohistochemical staining shows strong Nav1.5 expression in peripheral nervous tissue. Cardiac tis-sue was stained with Nav1.5 antibody in the absence (A) or presence (B) of a blocking peptide to denote positive and negative controls, respectively. Nav1.5 immunoreactivity was strongly exhibited in peripheral nerves (C), dorsal root ganglia (D), and to a lesser extent in the brain (E). Nav1.7 immunostaining was also performed on peripheral nerves (F), dorsal root ganglia (G), and brain (H) for comparison.

Table I.—�Candidate gene variants segregating in affected family members.chr# Position gene symbol Protein Variant gene region translation impact dbSNP ID (rs#)

2 234599422 UGT1A6; UGT1A7; UGT1A8; UGT1A9

Promoter; Intronic 17862860


Promoter; Intronic


Promoter; Intronic

3 38645378 SCN5A p.A572D exonic Missense 362104236 46563804 CYP39A1 p.K329Q; p.K157Q;

p.K309Qexonic Missense 41273654

7 99358524 CYP3A4 p.M444T; p.M445T exonic Missense 498691021 35742799 KCNE2 p.t8a exonic Missense 2234916



demonstrated resistance to at least one la fol-lowing subcutaneous infiltration, with 36% re-sponding only to mepivicaine and 17% only to lidocaine.10 all available las bind to sodium channel subtypes indiscriminately, and differ-ential blockade is not observed clinically. the la binding site does not differ substantially between sodium channel subtypes, and LAs have multiple points for molecular rotation, which may limit specificity for individual sub-types.21

our results suggest that a variant form of the SCN5A gene encoding for Nav1.5 is associated with la resistance in this family. Nav1.5 has been extensively studied for its action in car-diomyocytes, but its role in the nervous system is unknown. We noted that Nav1.5 is detectable by immunohistochemistry in postmortem hu-man samples of peripheral nerves, dorsal root ganglia, and cerebral cortex. Microarray data from the allen Brain atlas show that SCN5A (Nav1.5) and SCN9A (Nav1.7) are expressed in the human cerebral cortex and subcortical structures.20

The A572D variant in SCN5A is present in 0.5% of the general population and was initially identified in an individual with long Qt syndrome.22 Of note, the proband did not meet diagnostic criteria for long Qt syndrome because her EKG demonstrated normal sinus rhythm and right ventricular conduction de-lay with a Qt interval of 400 msec and a Qtc of 444 msec despite her carrying the A572D variant. Subsequent studies noted an associa-tion between the variant and sudden cardiac death and long QT syndrome; however, fur-ther study demonstrated that A572D was not causative in either condition.23 this variant results in an amino acid change from alanine to aspartate at position 572 located between the D1-D2 inter-domain linker of the Nav1.5 sodium channel.22 A572D confers a suscepti-bility to arrhythmia when present in cardio-myoctes by mimicking phosphorylation at the S571 regulatory site on Nav1.5 that is targeted by CAMKII, decreasing the resting membrane potential, increasing the probability of depo-larization, and resulting in a shorter recovery time compared to the wild-type sodium chan-

throughout the cell body, along the peripheral nerve fiber and on dorsal root ganglia. Nav1.7 immunostaining served as a positive control for peripheral nerve and dorsal root ganglia. these results demonstrate that Nav1.5 is ex-pressed in peripheral nerves, suggesting this channel may be involved in pain perception, similar to Nav1.7.

to investigate whether Nav1.5 is present in the central nervous system, we performed im-munohistochemistry on healthy human cere-bral cortex and spinal cord. the results were compared to the presence of Nav1.7 by im-munohistochemistry. Nav1.5 and Nav1.7 were detected on healthy human cerebral cortex but not spinal cord. Microarray expression data from six healthy human brains demonstrate that Nav1.5 and Nav1.7 transcripts are present in significant quantities, corroborating our im-munohistochemical data.20

Discussion

resistance to la has been documented in humans, and prior exposure to scorpion ven-om and genetic variance are putative mecha-nisms.5-7, 10 During routine clinical care, we identified a patient demonstrating a lack of lo-cal anesthesia with lidocaine, mepivicaine, and ropivacaine and a lack of motor blockade with mepivacaine and ropivacaine.

to explore whether la resistance is asso-ciated with genetic variation in voltage-gated sodium channels, we performed whole-exome sequencing on a family with three of four indi-viduals reporting LA resistance. Patients with ehlers-Danlos syndrome type iii consistently demonstrate la resistance as determined by pain threshold to laser stimulation following topical or subcutaneous la administration.8 this resistance is not due to rapid dispersion of the medication as measured by movement of radiolabelled solution in ehlers-Danlos syn-drome type iii following deep dermal injec-tion, suggesting that the LA fails for another reason.9 genetic variation may account for the variability in sensitivity to different las iden-tified in a case series of patients self-reporting LA resistance. Of the 250 patients tested, 53%



and may be affected. genetic testing is cost prohibitive, with clinical exome sequncing costing less than $6000 per test and clinical single nucleotide polymorphism testing cost-ing approximately $200. a diagnosis of local anesthetic resistance is based on clinical find-ings, but validation of a genetic cause of local anesthetic resistance would permit confimatory testing in the future. Whole-exome sequencing of a family with la resistance revealed that the affected individuals shared an A572D mutation on Nav1.5. Although Nav1.5 is the predomi-nant sodium channel in cardiomyocytes, it is also detectable on human peripheral nerves, dorsal root ganglia, and cerebral cortex. These results provide evidence for a possible genetic etiology for LA resistance in humans, which warrants further research. Clinically, a failed regional block should prompt a screening test with skin infiltration of multiple local anes-thetics followed by sensory assessment. if one or more local anesthetics result in loss of skin sensation the regional block should be repeated with the effective local anesthetic. if none of the local anesthetics are effective, they will re-quire general anesthesia for the scheduled pro-cedure. the patient should be informed about their complete local anesthetic resistance and have the diagnosis documented in their medi-cal record.

Conclusions

resistance to local anesthetics in humans has been documented in the literature, and one postulated mechanism is related to ge-netic variance. We identified a genetic vari-ant in the gene encoding for Nav1.5 that is associated with la resistance. We also dem-onstrate that Nav1.5 is present in human pe-ripheral nerves to support the plausibility that an abnormal form of the Nav1.5 protein could be responsible for the observed la resistance. limitations of this study are that only one family with local anesthetic resis-tance was studied and the mutation identified may be correlated with local anesthetic resis-tance without being causative. Future work is needed to test for genetic variants in more

nel.24, 25 CAMKII also regulates membrane resting potential in neurons.26 We postulate that the A572D Nav1.5 mutation results in in-creased probability of membrane depolariza-tion in peripheral nerves despite exposure to las.

Furthermore, point mutations in both SC-N5A (Nav1.5) and SCN9A (Nav1.7) have been shown to decrease the effect of la on sodium channels. The F1486 deletion in Nav1.5 abol-ishes binding of lidocaine to the channel in the inactivated state, resulting in resistance to lidocaine’s blockade of the channel.27 the F1737A mutation in Nav1.7 results in de-creased affinity of lidocaine and tetracaine for the resting state of the Nav1.7 channel and decreases use-dependent block of the chan-nel by both las.28 Both of these mutations are near the pore region, allowing sodium ion conductance.29 In contrast, the mutation we identified is distant from the pore region and la binding site on sodium channels and may confer la resistance by altering rest-ing membrane potential rather than interfer-ing with drug binding. Calcium, calmodulin, and CAMKII have been shown to alter car-diac and neuronal membrane excitability by regulating sodium channel inactivation and manifesting clinically as arrhythmia or au-tism.30, 31 Similarly, the A572D mutation on Nav1.5 may interfere with calcium-mediated regulation and manifest as clinically observed la resistance.

KCNE2

the family members demonstrating la resistance also shared a mutation in KCNE2, which encodes for a potassium ion channel in-hibited by LAs (LQT5).32 although KCNE2 is widely expressed in adult mouse spinal cord as noted in the Allen Brain Atlas,33 the Human Protein Atlas does not demonstrate the pres-ence of significant LQT5 in human peripheral nerves.34 Thus, the observed KCNE2 mutation is not likely to cause la resistance.

The frequency of local anesthetic resistance is unknown, but approximately 36 million people worldwide carry the A572D mutation



zygous triadin Mutations associated With autosomal-Recessive Long-QT Syndrome and Pediatric Sudden Cardiac Arrest Elucidation of the Triadin Knockout Syn-drome. Circulation 2015;131:2051-60.

12. Babraham Bioinformatics; [Internet]. Available from: http://www.bioinformatics.babraham.ac.uk/projects/fastqc/ [cited 2016, Aug 31].

13. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. 2009;25:2078-9.

14. McKenna A, Hanna M, Banks E, Sivachenko A, Cibul-skis K, Kernytsky A, et al. The Genome Analysis Toolkit: a Mapreduce framework for analyzing next-generation DNA sequencing data. Genome Res 2010;20:1297-303.

15. SnpEff; [Internet]. Available from: http://snpeff.source-forge.net/ [cited 2016, Aug 31].

16. Flanagan SE, Patch A-M, Ellard S. Using SIFT and PolyPhen to predict loss-of-function and gain-of-function mutations. Genet Test Mol Biomarkers 2010;14:533-7.

17. IGSR: The International Genome Sample Resource; [In-ternet]. Available from: http://www.1000genomes.org/ [cited 2016, Aug 31].

18. SIFT; [Internet]. Available from: http://sift.jcvi.org [cited 2016, Aug 31].

19. PolyPhen-2 prediction of functional effects of human ns-SNP; [Internet]. Available from: http://genetics.bwh.har-vard.edu/pph2/ [cited 2016, Aug 31].

20. Hawrylycz MJ, Lein ES, Guillozet-Bongaarts AL, Shen EH, Ng L, Miller JA, et al. an anatomically comprehen-sive atlas of the adult human brain transcriptome. Nature 2012;489:391-9.

21. Chevrier P, Vijayaragavan K, Chahine M. Differential modulation of Nav1.7 and Nav1.8 peripheral nerve so-dium channels by the local anesthetic lidocaine. Br J Pharmacol 2004;142:576-84.

22. Tester DJ, Valdivia C, Harris-Kerr C, Alders M, Salis-bury BA, Wilde AA, et al. Epidemiologic, molecular, and functional evidence suggest A572D-SCN5A should not be considered an independent LQT3-susceptibility muta-tion. Heart Rhythm 2010;7:912-9.

23. Milan DJ, Melman YF, Ellinor PT. Rare ion channel pol-ymorphisms: separating signal from noise. Heart Rhythm 2010;7:920-1.

24. Koval OM, Snyder JS, Wolf RM, Pavlovicz RE, Glynn P, Curran J, et al. Ca2+/calmodulin-dependent protein ki-nase II-based regulation of voltage-gated Na+ channel in cardiac disease. Circulation 2012;126:2084-94.

25. Albert CM, Nam EG, Rimm EB, Jin HW, Hajjar RJ, Hunter DJ, et al. cardiac sodium channel gene vari-ants and sudden cardiac death in women. circulation 2008;117:16-23.

26. Hund TJ, Koval OM, Li J, Wright PJ, Qian L, Snyder JS, et al. A beta(IV)-spectrin/CaMKII signaling complex is essential for membrane excitability in mice. J Clin Invest 2010;120:3508-19.

27. Song W, Xiao Y, Chen H, Ashpole NM, Piekarz AD, Ma P, et al. The human Nav1.5 F1486 deletion associ-ated with long Qt syndrome leads to impaired sodium channel inactivation and reduced lidocaine sensitivity. J Physiol 2012;590:5123-39.

28. Panigel J, Cook SP. A point mutation at F1737 of the hu-man Nav1.7 sodium channel decreases inhibition by local anesthetics. J Neurogenet 2011;25:134-9.

29. scholz a. Mechanisms of (local) anaesthetics on volt-age-gated sodium and other ion channels. Br J Anaesth 2002;89:52-61.

30. Kim J, Ghosh S, Liu H, Tateyama M, Kass RS, Pitt GS. Calmodulin mediates Ca2+ sensitivity of sodium chan-nels. J Biol Chem 2004;279:45004-12.

31. Weiss LA, Escayg A, Kearney JA, Trudeau M, Mac-Donald BT, Mori M, et al. Sodium channels SCN1A,

patients with local anesthetic resistance and the variant sodium channels will need to be tested in vitro to confirm the local anesthetic resistance.

Key messages

— complete local anesthetic resistance is uncommon but may be diagnosed by confirming medication delivery around a nerve and testing for loss of sensation or motor function.

— the Nav1.5 sodium channel is pres-ent on both cardiomyocytes and peripheral nerves.

— a family with local anesthetic re-sistance shared the A572D mutation in the SCN5A gene which may indicate a genetic etiology for local anesthetic re-sistance.

References

1. Covino BG. Local anesthesia. N Engl J Med 1972;286:975-83.

2. Tsang SY, Tsushima RG, Tomaselli GF, Li RA, Backx PH. A multifunctional aromatic residue in the external pore vestibule of Na+ channels contributes to the local anesthetic receptor. Mol Pharmacol 2005;67:424-34.

3. Cahalan MD. Local anesthetic block of sodium channels in normal and pronase-treated squid giant axons. Biophys J 1978;23:285-311.

4. Tarkkila PJ. Incidence and causes of failed spinal anes-thetics in a university hospital: a prospective study. Reg Anesth 1991;16:48-51.

5. Kavlock R, Ting PH. Local anesthetic resistance in a pregnant patient with lumbosacral plexopathy. BMc an-esthesiol 2004;4:1.

6. Hoppe J, Popham P. Complete failure of spinal anaesthe-sia in obstetrics. Int J Obstet Anesth 2007;16:250-5.

7. Panditrao MM, Panditrao MM, Khan MI, Yadav N. Does scorpion bite lead to development of resistance to the effect of local anaesthetics? Indian J Anaesth 2012;56: 575-8.

8. Arendt-Nielsen L, Kaalund S, Bjerring P, Hogsaa B. In-sufficient effect of local analgesics in Ehlers Danlos type iii patients (connective tissue disorder). acta anaesthe-siol Scand 1990;34:358-61.

9. Oliver DW, Balan KK, Burrows NP, Hall PN. Dispersal of radioisotope labelled solution following deep dermal injection in Ehlers-Danlos syndrome. Br J Plast Surg 2000;53:308-12.

10. Trescot AM. Local anesthetic “resistance”. Pain Physi-cian 2003;6:291-3.

11. Altmann HM, Tester DJ, Will ML, Middha S, Evans JM, Eckloff BW, et al. Homozygous/Compound Hetero-



Conflicts of interest.—steven clendenen receives speaker’s fees from the institute for advanced Medical education and Mallinck-rodt Pharmaceuticals.Congresses.—Abstract presented at the 2015 Annual Meeting of the International Anesthesia Research Society Honolulu, Hawaii March 21-24, 2015.Acknowledgements.—The authors would like to thank Victoria L. Clifton, MLIS, ELS (Academic and Research Support, Mayo Clin-ic, Jacksonville, FL, USA), for her editorial assistance with the preparation of the manuscript and Ilana I. Logvinov, MSN, RN (As-sistant Professor of Nursing, Department of Anesthesiology, Mayo Clinic, Jacksonville, FL, USA), for her assistance with the study.Article first published online: June 1, 2016. - Manuscript accepted: May 30, 2016. - Manuscript revised: May 17, 2016. - Manuscript received: February 19, 2016.

33. Allen Spinal Cord Atlas. Allen Institute for Brain Sci-ence; 2015 [Internet]. Available from: http://mousespinal.brain-map.org [cited 2015, Jun 5].

34. Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oks-vold P, Mardinoglu A, et al. Proteomics. Tissue-based map of the human proteome. Science 2015;347:1260419.

SCN2A and SCN3A in familial autism. Mol Psychiatry 2003;8:186-94.

32. Friederich P, Solth A, Schillemeit S, Isbrandt D. Lo-cal anaesthetic sensitivities of cloned HERG channels from human heart: comparison with HERG/MiRP1 and HERG/MiRP1 T8A. Br J Anaesth 2004;92:93-101.


last years, introduction of high flow oxygen in the adult icU 1 has brought another option in the armamentarium of respiratory failure manage-ment. Until very recently, the exact place of this technique was not clearly defined, but thought somewhere between conventional oxygenation and NIV, without challenging the preexisting order in the ventilatory management of acute respiratory failure. changes in the management of patients with acute hypoxemic respiratory

acute respiratory failure and need of respi-ratory support remains one of the leading

causes of ICU admission. For patients with de novo (i.e., hypoxemic) acute respiratory fail-ure, respiratory support included until recently, conventional oxygenation, in some instances noninvasive ventilation (NIV) and ultimately invasive mechanical ventilation. During these

E X P E R T S ’ O P I N I O N

Hazards of intubation in the icU: role of nasal high flow oxygen therapy for preoxygenation

and apneic oxygenation to prevent desaturationJean-Damien ricarD 1, 2, 3

1aP-HP, Hôpital louis Mourier, service de réanimation Médico-chirurgicale, colombes, France; 2inserM, iaMe 1137, Paris, France; 3Université Paris Diderot, iaMe 1137, sorbonne Paris cité, F-75018 Paris, France*corresponding author: Jean-Damien ricard, service de réanimation Médico-chirurgicale, Hôpital louis Mourier, 178 rue des renouillers, F-92700 colombes, France. e-mail: [email protected]


lavoro: 11264-Mastitolo breve: HaZarDs oF intUBation in tHe icUprimo autore: ricarDpagine: 1098-106citazione: Minerva anestesiol 2016;82:1098-106

a B s t r a c tacute respiratory failure is one the most common motives for intensive care unit admission. although results from recent studies with high flow nasal oxygen have challenged our current management of these patients, a substantial number of them will require invasive mechanical ventilation and tracheal intubation. Life-threatening hypoxemia is the most frequent complication of these intubations. Desaturations occur despite properly conducted preoxygenation. Hence, alternatives are warranted to improve oxygenation during intubation. Two phases may be distinguished: the actual pre-oxygenation period (whilst the patient is still breathing spontaneously) and the laryngoscopy that requires rapid sequence induction. Noninvasive ventilation improves preoxygenation and limits desaturation but oxygen supply is interrupted to allow for laryngoscopy. High flow oxygen is increasingly used to manage patients with hypoxemic acute respiratory failure and can be maintained during the intubation procedure with the advantage of pursuing oxygen supply during pa-tient’s apnea, thereby providing apneic oxygenation. Discrepant results on the superiority of high flow oxygen compared to conventional facemask preoxygenation to limit desaturation during intubation highlight key determinants of effective apneic oxygenation: patent upper airway (importance of jaw thrust), and sufficient and constant administration of oxygen (high flows of 60 L/min rather than 15 L/min). Studies comparing high flow oxygen to noninvasive ventilation are ongo-ing and will help clarify the indications of each technique. This paper aims to show the evidence on the potential high flow nasal oxygen bears to improve preoxygenation for intubation outside the operating room. A practical algorithm to decide which preoxygenation device to use is proposed.(Cite this article as: ricard JD. Hazards of intubation in the ICU: role of nasal high flow oxygen therapy for preoxygenation and apneic oxygenation to prevent desaturation. Minerva Anestesiol 2016;82:1098-106)Key words: Respiration, artificial - Intubation - Respiratory insufficiency.


comment in p. 1036.

HaZarDs oF intUBation in tHe icU ricarD


erating room (OR), it is generally agreed upon that the major problem posed in the icU is oxygenation rather than intubation per se, as in the or.13

one of the most common features of critically ill patients’ intubation is occurrence of profound desaturation (defined as SpO2 below 80%). This has been observed in all the studies on the sub-ject with a striking consistency in the propor-tion of patients experiencing such desaturation, between 20 and 25%.14-16 one can intuitively suspect that the profounder the patient’s baseline hypoxemia, the greater the risk of desaturation during laryngoscopy. in addition, there is a di-rect relationship between the number of attempts and the incidence of desaturation, an intubation requiring more than two attempts increases by more than 10-fold the risk of severe hypox-emia.17 in turn, the greater the desaturation, the greater the risk for cardiac arrest.18 Hence, ensuring adequate oxygenation throughout the procedure is of utmost importance.

although it is beyond the scope of this re-view to discuss in detail this aspect, one must bear in mind that optimal timing of intubation is a crucial and key determinant in the risk of desaturation. indeed, one of the most challeng-ing decisions in the management of patients with hypoxemic acute respiratory failure is to decide when to move from spontaneous breath-ing to invasive mechanical ventilation.19 since it has been shown in the past that delaying in-tubation and invasive mechanical ventilation may worsen patient outcome, use of strict, simple, predefined criteria to initiate intubation are necessary to avoid detrimental delays. For example, the Florali study used the following criteria:2 1) signs of persisting or worsening respiratory failure, defined by at least two of the following criteria: a respiratory rate above 40 breaths/min, lack of improvement of signs of high respiratory-muscle workload, develop-ment of copious tracheal secretions, acidosis with a pH below 7.35, SpO2 below 90% for more than 5 min without technical dysfunction, or intolerance to oxygenation techniques; or 2) hemodynamic instability (defined by a systolic blood pressure below 90 mmHg, a mean blood pressure below 65 mmHg or requirement for

failure are however bound to occur following the resounding demonstration of high flow na-sal cannula oxygen’s efficacy in these patients.2 This study unambiguously showed that these patients had a lower mortality when treated with high flow nasal oxygen compared to those treated either with conventional oxygen or in combination with NIV. In addition, use of high flow nasal oxygen avoided intubation in the sub-group of patients exhibiting a low PaO2/Fio2 ratio (<200). These cogent results confirm those of a number of observational studies per-formed earlier that had showed that respiratory status of patients with acute respiratory failure was significantly improved with high flow na-sal oxygen.3-10 although they did suggest that intubation may have been avoided in some patients, design of these studies precluded any formal demonstration. nonetheless, and despite these changes, need for intubation remains im-portant when respiratory condition deteriorates and places these severely hypoxemic patients at higher risks of desaturation than patients with-out respiratory failure.

This review will focus on the potential high flow nasal oxygen bears to improve pre-oxygenation for intubation in the ICU setting. other aspects surrounding improvement of intubation will not be detailed and have been reviewed elsewhere.11

Hazards of intubation in the ICU

It has been long shown that intubation of critically ill patients in the icU is hazardous.12 this results from the combination of a multi-tude of factors (related to the patient, the envi-ronment and the situation) that all contribute to intubation’s dangerousness: unstable patients, rapidly deteriorating, with decreased physi-ologic reserve, full stomach and biological abnormalities that may worsen laryngoscopy feasibility because of increased risk of bleed-ing (coagulopathy, low platelet count, ongo-ing anticoagulants); physicians in charge with variable levels of expertise in airway manage-ment; and sometimes unanticipated extreme emergency. Although incidence of difficult intubation is higher in the icU than in the op-

ricarD HaZarDs oF intUBation in tHe icU


surgery, this phase enables to ensure several minutes of apnea without desaturation. In the ICU however, the instrumental studies by Tho-mas Mort have shown that, contrary to the OR, conventional preoxygenation is of limited ef-ficacy in critically ill patients. In one study, a total of 36% of patients had minimal changes (±5%) in their baseline PaO2, and only 19% increased their baseline Pao2 by at least 50 mmHg after properly conducted preoxygena-tion.28 In a subsequent study, he showed that prolonging preoxygenation from 4 to 8 min-utes provided little benefit in terms of oxygena-tion since Pao2 rose from to 83 to 92 mmHg.29 Numerous reasons concur to limit efficiency of preoxygenation in ICU patients: cardiopulmo-nary underlying disease, anemia, low cardiac output, hypermetabolic states, ventilation/per-fusion mismatch, obesity, pain, etc.

NIV has been to date, the most efficient means to improve preoxygenation. In the pivotal study by Baillard et al.,30 hypoxemic patients were randomized to receive either bagvalve preoxygenation or NIV for 4 min. in patients that received niv, spo2 was sig-nificantly higher than the control group at the end of the preoxygenation period. This ben-efit was maintained during laryngoscopy in most, but not all patients. importantly, several patients experienced profound desaturation (spo2<80%) during laryngoscopy despite the transient improvement obtained by niv dur-ing the preoxygenation period. This phenom-phenom-enon clearly underlines the necessity to ensure some form of oxygen supply during laryngos-copy, however short this act may be. By es-sence, this may not be accomplished by niv. In addition, a significant number of patients may not benefit from this technique because of contraindications (all the contraindications to niv, including neurological impairment, agitation, etc). Moreover, it has been shown that use of NIV for preoxygenation can induce gastric distension.31 although this is prob-ably inconsequential for patients in the or, it could be more problematic in higher-risk patients such as those in the ICU with a full stomach. Finally, the technique might be a lit-tle cumbersome to be applied systemically to

vasopressor); or 3) deterioration of neurologic status (Glasgow coma scale below 12 points).

Although high flow nasal oxygen may avoid further need for mechanical ventilation in some patients with acute respiratory failure,2, 8 it may unduly delay initiation of mechanical ventila-tion in others and worsen their outcome,20 as already evidenced for niv.21-25 Hence timely intubation is important to avoid morbidity and mortality associated with prolonged use of high flow oxygen. In that respect, it is in-teresting to note that in studies where high flow oxygen failure was not associated with increased mortality, duration of high flow was shorter in patients that failed high flow than in those who succeeded,6 whereas in studies in which high flow oxygen failure was associated with increased mortality, duration of high flow oxygen was significantly longer in case of fail-ure,26 highlighting the importance of judicious timing of intubation.27

Because high flow nasal oxygen has been used in severely hypoxemic patients, a number of them have naturally required invasive me-chanical ventilation given the high intubation rates in these patients. Hence, in our daily prac-tice we started keeping the high flow device in place while instituting the intubation proce-dure (rather than downgrading the oxygen sup-port to a facemask) and noticed remarkable in-tubation conditions.6 This prompted us, as well as others, to investigate the use of high flow oxygen to prevent desaturation during tracheal intubation of critically ill patients. there are several means by which high fl ow nasal oxy-high flow nasal oxy-gen may improve oxygenation during tracheal intubation and two distinct but contiguous phases of tracheal intubation, preoxygenation and perlaryngoscopy oxygenation need to be discussed. the former has received considera-ble attention (obviously in the or but also too in the ICU); which is less the case for the latter.

Means to improve preoxygenation

Preoxygenation, also termed denitrogena-tion, aims at bringing nitrogen level in the al-veolus to null with complete replacement by oxygen. In stable patients undergoing elective



reserve at the alveolar level to allow for the laryngoscopy to be performed without desatu-ration. While this is achieved in the majority of patients undergoing elective surgery, this is not the case in the critically ill.28, 29 Hence, ad-dition of pharyngeal oxygen via a small cath-eter has been suggested by some but without extensive evaluation in critically ill patients. In asa status 1 or 2 patients undergoing elective surgery, Teller et al evaluated the benefit of ad-ministering a 3-L oxygen flow via a nasal cath-eter during apnea. When patients received pha-ryngeal oxygen insufflation, saturation never fell below 97% during the entire 10-min apnea period and mean lowest oxygen saturation was 98%. In the absence of oxygen supplementa-tion, apnea was considerably shortened to 6.8 minutes and lowest mean oxygen saturation was only 91%.35 Similar findings were re-ported by taha et al. in asa 1 or 2 patients, randomized to receive either 5L/min oxygen via a nasopharyngeal catheter or no oxygen. in the latter group, spo2 fell to 95% within a mean apnea time of 3.65 minutes, whereas in the supplemental oxygen group, SpO2 was maintained in all patients at 100% throughout the 6 min of apnea.36 These results were re-produced in obese patients where nasal oxygen supplementation prolonged apnea time with-out desaturation.37 Of note, these studies were performed in the or, and their results may not be fully reproducible in the icU setting.

What is the purpose of administering oxy-gen at the pharyngeal level in the absence of respiratory movement? The answer lies in the concept of apneic oxygenation described decades ago by Draper et al.,38 and named at the time, diffusion respiration. During apnea, the constant circulation in the alveolar capil-laries of reduced hemoglobin drives oxygen absorption from the alveolus to the capillary blood at a rate of 250 mL oxygen/minute. This causes a decrease in alveolar pressure, and cre-ates a pressure gradient between the alveolus and the upper airway which generates a flow of air from the pharynx to the distal airway. If this air is enriched in oxygen, apneic oxygen-ation occurs. This phenomenon first described in dogs 38 was confirmed in patients undergo-

all patients requiring tracheal intubation in the ICU, as suggested by a follow-up study that showed that NIV was “spontaneously” used in less than half the patients requiring urgent intubation.32

Improvement of preoxygenation with high flow nasal cannula oxygen

High flow oxygen has clearly shown its su-periority over conventional facemask during acute respiratory failure to improve oxygena-tion. contributors to this improvement (high Fio2 along with high flow that enhances pha-ryngeal deadspace washout and also induces a modest PEEP effect that has been shown to increase end expiratory lung volume) could therefore also contribute to improve pre-oxygenation. In a quasi-experimental study, comparing preoxygenation performed “con-ventionally” with a nonrebreathing facemask (with the addition of 6 L/min oxygen flow ad-ministered through a nasopharyngeal catheter during the apnea period as routinely done in our ICU) to preoxygenation with high flow na-sal oxygen (60 L/min, FiO2:1), we were able to show that in patients with mild-to-moderate respiratory failure, oxygen saturation at the end of the preoxygenation period was signifi-cantly greater in patients that preoxygenated with high flow oxygen.33 This beneficial result with high flow was not found in another study using the same comparators and endpoints.34 in their study, vourc’h et al.34 observed similar saturation levels at the end of the preoxygena-tion period between high flow and conven-tional oxygen. Of note, the patients included in their study were more severe in terms of hy-poxemia. In addition, patients randomized to receive high flow as preoxygenation technique were possibly more severe since a greater pro-portion of them were already either under high flow or NIV.34

Means to improve perlaryngoscopy oxygenation

as mentioned above, the ultimate goal of preoxygenation is to provide sufficient oxygen



during the first minutes of the procedure in the high flow oxygen group, which was not the case in the conventional preoxygenation group, suggesting that some form of apneic oxygenation occurred. This was not the case in the Vourc’h study, since there was no differ-ence in the lowest saturation between the two groups (conventional preoxygenation and high flow oxygen).34 similarly, semler et al. found no difference between no oxygen and 15 L/min nasal oxygen during laryngoscopy in their so-called apneic oxygenation study.41 the pos-sible absence of two crucial determinants of apneic oxygenation may explain these discrep-ancies. First, one must bear in mind that jaw thrust is mandatory to ensure sufficient paten-cy of the upper airway, and avoid pharyngeal obstruction. Means to maintain patent upper airway are not mentioned, neither in Vourc’h

ing elective surgery. in a famous study, Fru-min et al. showed that patients could sustain a prolonged period of apnea (up to 55 min) without desaturating providing an adequate oxygen supply at the upper airway.39 Key de-terminants for effective apnea oxygenation to occur are the following: adequate circulation, adequate and constant supply of fresh oxygen, patent upper airway. Subsequent studies have highlighted some of these key factors. if, for example, oxygen is replaced by room air, then arterial oxygen partial pressure decreases very rapidly.40

table i summarizes the recent studies that aimed at preventing desaturation during la-ryngoscopy outside the or, all based on the principle of apneic oxygenation except for the use of niv in the study by Baillard et al.30 in our study,33 oxygen saturation remained stable

Table I.—�Interventions to prevent desaturation during emergent intubation outside the operating room.

Study first authors, year of publication

study design, setting, n. of patients

comparators (number of patients per

group)Main outcome measure

number of patients with desaturation

<80% (or % of pts)

comments

Baillard,30 2005

open label, randomized trial

icU, 53 patients

– NIV (26) – Bagvalve (27)

Lowest SpO2:NIV: 93%Bagvalve: 81%

niv: 2Bagvalve: 12

All patients with NIV contraindications excluded; no oxygen supply during laryngoscopy

Miguel-Montanes,33

2015Quasi-experimental

sequential study,icU, 101 patients

– High flow nasal o2 (51)

– High Fio2 facemask (50)

Lowest SpO2:High flow: 99%Facemask: 94%

High flow: 1Facemask: 7

non randomized, severe hypoxemic patients not included (already preoxygenated with high flow nasal O2)

vourc’h,34

2015rcticU, 119 patients

– High flow nasal o2 (62)

– High Fio2 facemask (57)

Lowest SpO2High flow: 91.5Facemask: 89.5

High flow: 16Facemask: 13

no evidence of upper airway patency (no mention of jaw thrust)

Wimalasena,43

2015retrospective,

before afterout-of-hospital,

728 patients

– nasal o2 (15L/min) (310)

– No oxygen (418)

% patients with SpO2 below 93% (any time):

nasal o2: 16.5No oxygen: 22.6

retrospective, recordings of spo2 were reported by clinicians (no direct readings from stored data)

semler,41 2015

open label, randomized trial

icU, 150

– nasal o2 (15L/min) (77)

– No oxygen (73)

Lowest SpO2nasal o2: 92%no o2: 90%

nasal o2: 12no o2: 18

no evidence of upper airway patency (no mention of jaw thrust); not really “high” flow (only 15 L/min)

sakles,42 2016

observational before, after study, emergency depart-ment, 127

– nasal o2 (15L/min) (72)

– No oxygen (55)

% patients with SpO2 below 90% – nasal o2: 7% – no o2: 29%

nasal o2: 4%no o2:18%



limits of NIV and high flow nasal oxygen when used to preoxygenate patients in the ICU.

Future studies

As mentioned above, NIV has been shown to improve preoxygenation in patients with acute respiratory failure. Such improvement was also seen in some but not all studies using high flow nasal oxygen in comparison with standard pre-oxygenation. Given the recent results regard-ing high flow nasal oxygen and NIV in patients with acute respiratory failure, the question arises as to how high flow oxygen compares with NIV to prevent desaturation during pre-oxygenation in patients with acute hypoxemic respiratory failure. a large multicenter rct (Florali2) will randomize patients who require urgent intubation in the ICU either to high flow nasal oxygen or to NIV to ensure preoxygen-ation (NCT02668458). Another study, compar-ing use of NIV alone to NIV plus high flow to ensure oxygenation during intubation is cur-rently recruiting patients (NCT02530957).

Take home message

A practical approach to preoxygenation is provided in Figure 1, based on patient severity and type of oxygen support present at the time

et al.,34 nor in the study by semler et al.41 Jaw thrust was not part of their procedure, or at least not mentioned as such. one can therefore hypothesize that a certain degree of pharyn-geal obstruction may have occurred, explain-ing why high flow oxygen was not superior to either low flow, or no oxygen. Second, Semler et al. used only 15 L/min oxygen in their in-terventional group. This is clearly not “high” flow, in comparison with the 60 L/min that can be achieved with the high flow nasal cannula devices. Again, combined with the absence of jaw thrust, one can easily imagine that apneic oxygenation did not occur. Outside the ICU, two studies 42, 43 recently reported results of the implementation of apneic oxygenation in a rapid sequence induction intubation protocol (Table I). Both studies found that implement-ing 15 L/min supplemental oxygen via nasal cannula during laryngoscopy significantly re-duced incidence of desaturation during intuba-tion of either intracranial haemorrhage patients in the emergency department,42 or of out-of-hospital patients.43 these results contradict those of semler et al.41 despite the use of the same oxygen flow. Taken together, there does seem to exist a benefit in apneic oxygenation and in the use of high flow oxygen. Further studies are warranted to confirm these benefi-cial effects. table ii summarizes strengths and

Table II.—�Comparison of NIV and high flow nasal oxygen during preoxygenation.strength limits settings

noninvasive ventilation – Positive pressure allowing alveolar recruitment

– strong evidence from one randomized study

– several contraindications (coma, agitation, uncooperative patient, contraindications to NIV)

– risk of gastric distension – No oxygen during

laryngoscopy – results not replicated – generalizability?

– Pressure support: 5-15 cmH2O (target expired tidal volume 6-8 mL/kg), PEEP: 5cmH2o, Fio2:1

High flow nasal oxygen – simplicity of use – Feasibility – Allows pursuit of

oxygenation during laryngoscopy providing apneic oxygenation

– same device kept from patient admission to intubation

– Discrepant results between studies

– limited alveolar recruitment

– Flow: 60 L/min – Fio2: 1



icU. these desaturations occur during apnea and despite properly preoxygenation. Thus im-proving perlaryngoscopy oxygenation is a cru-cial step to reduce morbidity of urgent tracheal intubation in the ICU. Because high flow oxy-gen is becoming the first line ventilatory sup-port for acute respiratory failure, it should be initiate right from icU admission (and perhaps before in the ED) and for practical reasons, can be easily maintained if patient requires intuba-tion. Future studies will provide insight into which will require NIV, and those that might even require both: preoxygenation with NIV and apneic oxygenation with high flow oxygen.

of the decision to intubate the patient. although one may argue that patients with acute respira-tory failure should no longer be managed with a high Fio2 facemask given the accumulating data showing the superiority of high flow oxy-gen, we decided to keep the facemask in the algorithm because indication for intubation may be governed by non-respiratory reasons. We also acknowledge the fact that there is dis-crepant findings regarding high flow oxygen, at least in the more hypoxemic patients, and that to date, no study has compared high flow oxygen to NIV. Hopefully, this algorithm will be simplified when the above-mentioned stud-ies will have been conducted.

Conclusions

Life threatening hypoxemia is the most fre-quent reported complication of intubation in the

Figure 1.—Pragmatic algorithm of intubation in the icU.1niv: noninvasive ventilation; 2using the MacocHa scale;44 3except in case of left heart failure or cardiogenic pulmonary edema.

Key messages

— life-threatening desaturation is a frequent complication of intubation in criti-



10. Frat J-P, Brugiere B, ragot s, chatellier D, veinstein a, goudet v, et al. Sequential application of oxygen therapy via high-flow nasal cannula and noninvasive ventilation in acute respiratory failure: an observational pilot study. respir care 2015;60:170-8.

11. De Jong a, Jung B, Jaber s. intubation in the icU: we could improve our practice. Crit Care Lond Engl 2014;18:209.

12. Schwartz DE, Matthay MA, Cohen NH. Death and other complications of emergency airway management in criti-cally ill adults. a prospective investigation of 297 trache-al intubations. anesthesiology 1995;82:367-76.

13. Isono S, Ishikawa T. Oxygenation, not intubation, does matter. anesthesiology 2011;114:7-9.

14. Jaber s, amraoui J, lefrant JY, arich c, cohendy r, landreau l, et al. clinical practice and risk factors for immediate complications of endotracheal intubation in the intensive care unit: a prospective, multiple-center study. crit care Med 2006;34:2355-61.

15. Simpson GD, Ross MJ, McKeown DW, Ray DC. Tra-cheal intubation in the critically ill: a multi-centre na-tional study of practice and complications. Br J anaesth 2012;108:792-9.

16. griesdale De, Bosma tl, Kurth t, isac g, chittock Dr. complications of endotracheal intubation in the critically ill. intensive care Med 2008;34:1835-42.

17. Mort tc. emergency tracheal intubation: complications associated with repeated laryngoscopic attempts. Anesth analg 2004;99:607-13.

18. Mort tc. the incidence and risk factors for cardiac arrest during emergency tracheal intubation: a justification for incorporating the asa guidelines in the remote location. J clin anesth 2004;16:508-16.

19. laghi F, tobin M. indications for mechanical ventilation. in: Principles and Practice of mechanical ventilation. Third edition. New York: McGraw Hill; 2013. p. 101-35.

20. Kang BJ, Koh Y, lim c-M, Huh JW, Baek s, Han M, et al. Failure of high-flow nasal cannula therapy may delay intubation and increase mortality. intensive care Med 2015;41:623-32.

21. Moretti M, cilione c, tampieri a, Fracchia c, Marchioni a, nava s. incidence and causes of non-invasive me-chanical ventilation failure after initial success. Thorax 2000;55:819-25.

22. Agarwal R, Aggarwal AN, Gupta D. Role of noninvasive ventilation in acute lung injury/acute respiratory dis-tress syndrome: a proportion meta-analysis. respir care 2010;55:1653-60.

23. antonelli M, conti g, Moro Ml, esquinas a, gonzalez-Diaz g, confalonieri M, et al. Predictors of failure of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. intensive care Med 2001;27:1718-28.

24. esteban a, Frutos-vivar F, Ferguson nD, arabi Y, apez-teguía c, gonzález M, et al. noninvasive positive-pres-sure ventilation for respiratory failure after extubation. N engl J Med 2004;350:2452-60.

25. carrillo a, gonzalez-Diaz g, Ferrer M, Martinez-�uin-carrillo a, gonzalez-Diaz g, Ferrer M, Martinez-�uin-tana Me, lopez-Martinez a, llamas n, et al. non-invasive ventilation in community-acquired pneumonia and severe acute respiratory failure. intensive care Med 2012;38:458-66.

26. Kang BJ, Koh Y, lim cM, Huh JW, Baek s, Han M, et al. Failure of high-flow nasal cannula therapy may delay intubation and increase mortality. intensive care Med [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25691263 [cited 2015, Feb 18].

27. ricard JD, Messika J, sztrymf B, gaudry s. impact on outcome of delayed intubation with high-flow nasal can-nula oxygen: is the device solely responsible? Intensive care Med 2015;41:1157-8.

References

1. Ricard J-D. The high flow nasal oxygen in acute respira-tory failure. Minerva anestesiol 2012;78:836-41.

2. Frat J-P, thille aW, Mercat a, girault c, ragot s, Per-bet s, et al. High-flow oxygen through nasal cannula in acute hypoxemic respiratory failure. N Engl J Med 2015;372:2185-96.

3. sztrymf B, Messika J, Bertrand F, Hurel D, leon r, Dreyfuss D, et al. Beneficial effects of humidified high flow nasal oxygen in critical care patients: a prospective pilot study. intensive care Med 2011;37:1780-6.

4. lenglet H, sztrymf B, leroy c, Brun P, Dreyfuss D, Ricard J-D. Humidified high flow nasal oxygen during respiratory failure in the emergency department: feasibil-ity and efficacy. Respir Care 2012;57:1873-8.

5. sztrymf B, Messika J, Mayot t, lenglet H, Dreyfuss D, Ricard JD. Impact of high-flow nasal cannula oxygen therapy on intensive care unit patients with acute respi-ratory failure: a prospective observational study. J crit care 2012;27:324.e9-13.

6. Messika J, Ben ahmed K, gaudry s, Miguel-Montanes r, rafat c, sztrymf B, et al. Use of high-flow nasal can-nula oxygen therapy in subjects with ards: a 1-year obser-vational study. respir care 2015;60:162-9.

7. rello J, Perez M, roca o, Poulakou g, souto J, laborda c, et al. High-flow nasal therapy in adults with severe acute respiratory infection: a cohort study in patients with 2009 influenza A/H1N1v. J Crit Care 2012;27:434-9.

8. roca o, de acilu Mg, caralt B, sacanell J, Masclans Jr. Humidified high flow nasal cannula supportive therapy improves outcomes in lung transplant recipients readmit-ted to the intensive care unit because of acute respiratory failure. transplantation 2015;99:1092-8.

9. roca o, Perez-teran P, Masclans Jr, Perez l, galve e, evangelista a, et al. Patients with New York Heart Asso-Patients with New York Heart Asso-ciation class III heart failure may benefit with high flow nasal cannula supportive therapy: high flow nasal can-nula in heart failure. J crit care 2013;28:741-6.

cally ill patients in the intensive care unit despite appropriate preoxygenation.

— Alternatives to conventional preoxy-genation are warranted to prevent these profound desaturations from occurring and noninvasive ventilation and high flow nasal oxygen may contribute to improve preoxy-genation and limit these desaturations.

— High flow nasal oxygen has the ad-ditional benefit of providing apneic oxy-genation during rapid sequence induction although discrepant results have been re-ported that highlight the determinants of apneic oxygenation.

— When using high flow nasal oxygen to preoxygenate patients requiring intuba-tion, jaw thrust must be performed to en-sure patent airway and high oxygen flow rates should be used.



Hakki MA, Baraka AS. Nasopharyngeal oxygen insuf-flation following pre-oxygenation using the four deep breath technique. anaesthesia 2006;61:427-30.

37. Ramachandran SK, Cosnowski A, Shanks A, Turner CR. Apneic oxygenation during prolonged laryngoscopy in obese patients: a randomized, controlled trial of nasal oxygen administration. J Clin Anesth 2010;22:164-8.

38. Draper WB, richard W. Whitehead. Diffusion respiration in the dog anesthetized by pentothal sodium. anesthesiol-ogy 1944;5:262-73.

39. Frumin MJ, Epstein RM, Cohen G. Apneic oxygenation in man. anesthesiology 1959;20:789-98.

40. Heller ML, Watson TR, Imredy DS. Apneic oxygenation in man: polarographic arterial oxygen tension study. An-esthesiology 1964;25:25-30.

41. Semler MW, Janz DR, Lentz RJ, Matthews DT, Norman Bc, assad tr, et al. Randomized Trial of Apneic Oxy-genation during endotracheal intubation of the critically ill. am J respir crit care Med 2016;193:273-80.

42. Sakles JC, Mosier JM, Patanwala AE, Dicken JM. Ap-neic oxygenation is associated with a reduction in the incidence of hypoxemia during the RSI of patients with intracranial hemorrhage in the emergency department. intern emerg Med 2016;11:983-92.

43. Wimalasena Y, Burns B, reid c, Ware s, Habig K. Apneic oxygenation was associated with decreased de-saturation rates during rapid sequence intubation by an australian helicopter emergency medicine service. ann emerg Med 2015;65:371-6.

44. De Jong a, Molinari n, terzi n, Mongardon n, arnal JM, guitton c, et al. Early identification of patients at risk for difficult intubation in the intensive care unit: development and validation of the MacocHa score in a multicenter cohort study. am J respir crit care Med 2013;187:832-9.

28. Mort TC. Preoxygenation in critically ill patients re-quiring emergency tracheal intubation. crit care Med 2005;33:2672-5.

29. Mort TC, Waberski BH, Clive J. Extending the pre-oxygenation period from 4 to 8 mins in critically ill pa-tients undergoing emergency intubation. crit care Med 2009;37:68-71.

30. Baillard c, Fosse JP, sebbane M, chanques g, vincent F, courouble P, et al. noninvasive ventilation improves pre-noninvasive ventilation improves pre-oxygenation before intubation of hypoxic patients. Am J respir crit care Med 2006;174:171-7.

31. Delay JM, sebbane M, Jung B, nocca D, verzilli D, Pou-Delay JM, sebbane M, Jung B, nocca D, verzilli D, Pou-zeratte Y, et al. the effectiveness of noninvasive posi-the effectiveness of noninvasive posi-tive pressure ventilation to enhance preoxygenation in morbidly obese patients: a randomized controlled study. anesth analg 2008;107:1707-13.

32. Jaber s, Jung B, corne P, sebbane M, Muller l, chanques g, et al. an intervention to decrease complications re-lated to endotracheal intubation in the intensive care unit: a prospective, multiple-center study. intensive care Med 2010;36:248-55.

33. Miguel-Montanes r, Hajage D, Messika J, Bertrand F, gaudry s, rafat c, et al. Use of high-flow nasal cannula oxygen therapy to prevent desaturation during tracheal intubation of intensive care patients with mild-to-moder-ate hypoxemia. Crit Care Med 2015;43:574-83.

34. vourc’h M, asfar P, volteau c, Bachoumas K, clavieras n, egreteau P-Y, et al. High-flow nasal cannula oxygen during endotracheal intubation in hypoxemic patients: a randomized controlled clinical trial. intensive care Med 2015;41:1538-48.

35. Teller LE, Alexander CM, Frumin MJ, Gross JB. Pharyn-Teller LE, Alexander CM, Frumin MJ, Gross JB. Pharyn-Pharyn-geal insufflation of oxygen prevents arterial desaturation during apnea. anesthesiology 1988;69:980-2.

36. taha sK, siddik-sayyid sM, el-Khatib MF, Dagher cM,

Acknowledgements.—We wish to thank Dr Jonathan Messika for help and critical assessment of the manuscript.Conflicts of interests.—Jean-Damien Ricard received travel expenses coverage from Fisher&Paykel.Article first published online: May 7, 2016. - Manuscript accepted: May 4, 2016. - Manuscript revised: April 19, 2016. - Manuscript received: February 11, 2016.


in patients with acute kidney injury (aKi).4-10 Furthermore, the use of some kinds of fluids (particularly synthetic colloids), regardless of the volume, has been shown to increase both mortality 11 and the risk for renal replacement therapy (rrt) need (table i).12-16 Finally, it has long been known that colloids cause a dose-dependent impairment of coagulation, al-though to a different extent depending on the different molecules.17

The type of fluids used in the ICU varies widely among different countries. Until a few years ago, in europe colloids were used for fluid resuscitation much more than in USA and

Fluid therapy has always been considered a cornerstone of perioperative and critical

care medicine. Doubtless, hypovolemia and the consequent end-organ hypoperfusion can lead to fatal complications in surgical and/or intensive care unit (ICU) patients and, accord-ingly, are to be avoided. on the other hand, there is increasing evidence that an excessive administration of fluids may negatively affect survival in critically ill patients. Fluid over-load was associated with increased mortality in different ICU settings 1-3 and, in particular,


the risk of infusing gelatin? Die-hard misconceptions and forgotten (or ignored) truths

antonio PisaNo 1 *, giovanni laNDoNi 2, 3, rinaldo BelloMo 4, 5

1Cardiac Anesthesia and Intensive Care Unit, A.O.R.N. “Dei Colli”, Monaldi Hospital, Naples, Italy; 2Department of Anesthesia and Intensive Care, IRCCS San Raffaele Scientific Institute, Milan, Italy; 3Vita-salute san raffaele University, Milan, Italy; 4Department of Intensive Care, Austin Hospital, Melbourne, Australia;5Department of epidemiology and Preventive Medicine, australian and New Zealand intensive care research centre (aNZic-RC), Monash University, Melbourne, Australia.*corresponding author: antonio Pisano, via cupa della torretta, 20, 80070 Bacoli (Na), italy. e-mail: [email protected]


lavoro: 11129-Mastitolo breve: THE RISK OF INFUSING GELATINprimo autore: PisaNopagine: 1107-14citazione: Minerva Anestesiol 2016;82:1107-14

a B s t r a c tFluid therapy is considered a cornerstone of perioperative and critical care medicine. However, the type of fluids used varies widely among different countries. synthetic colloids may negatively affect coagulation and are potentially neph-rotoxic. “Modern” hydroxyethyl starches (HES) were widely used until recently when their association to mortality and renal replacement therapy risk among critically ill patients brought to restriction by the european Medicines agency in 2013. since synthetic colloids are traditionally thought to be much more effective as volume expanders than crystalloids (although their efficacy is probably largely overrated), there is concern that physicians, practically deprived of HES, could be tempted to rely again on “old” gelatins. The aim of this contribution is to warn clinicians that gelatins share all potential adverse effects of other synthetic colloids, and are possibly even more nephrotoxic than HES. Moreover, gelatins have no beneficial effects on outcomes as compared with crystalloids (on the contrary, they might even increase mortality), and are also more expensive. Accordingly, a “return” to gelatins should be strongly discouraged.(Cite this article as: Pisano a, landoni g, Bellomo r. the risk of infusing gelatin? Die-hard misconceptions and forgotten (or ignored) truths. Minerva Anestesiol 2016;82:1107-14)Key words: Fluid therapy - colloids - gelatin - acute kidney injury - renal replacement therapy.

comment in p. 1039.

Minerva Anestesiologica 2016 October;82(10):1107-14© 2016 eDiZioNi MiNerVa MeDicaonline version at http://www.minervamedica.it

PisaNo THE RISK OF INFUSING GELATIN


A sudden bereavement: the “hydroxyethyl starch” affair

in June 2013, the Pharmacovigilance and risk assessment committee (Prac) of the european Medicines agency (eMa) recom-mended the suspension of marketing authori-zations for all HES solutions.19 Not to mention the fact that most studies who guaranteed the safety of HES were fraudulent,20, 21 and that there were already several clear reports of re-nal toxicity 13, 22 and increased mortality 22 with HES use, this recommendation was mostly based on two large randomized controlled tri-als (rcts) published in 2012.11, 12 In the “6S” study, Perner et al.11 randomized 804 patients with severe sepsis to receive either 6% HES 130/0.42 or Ringer’s acetate (RA) for fluid resuscitation, and found a significant increase in 90-day mortality in the HES group (rela-tive risk [RR] 1.17; 95% confidence interval

New Zealand, where the use of crystalloids is largely predominant. Moreover, hydroxyethyl starches (HES), a family of modified maize- or potato-derived polysaccharides which vary in their molecular weight and mean number of hydroxyethyl residues per glucose unit (the so-called molar substitution), were the most used among colloids in some european coun-tries, including France, Denmark and italy, and were also widely used, along with gelatins, in other countries such as the United Kingdom and germany.18 in the last few years, how-ever, HES solutions were withdrawn from the market and then re-introduced with many re-strictions. Fearing that clinicians could replace HES with gelatins for those clinical conditions for which HES solutions are now contrain-dicated, we want to warn them that the same safety concerns that led to restrict the use of HES are likely to apply also to gelatins, the use of which should be accordingly discouraged.

Table I.—�Main clinical investigations reporting major adverse effects of either hydroxyethyl starches (HES) or gelatins. Crystalloid/colloid ratios are also indicated when available.

study type of study Population N. colloid control crystalloid/ colloid ratio

adverse outcomes

transfusions renal Mortality

Brunkhorst et al. (ViseP)13 rct severe sepsis 537 HES 200/0.5 ringer lactate 1.3 - Higher AKI rate (34.9 vs. 22.8%, P=0.002)

andrrt need

(31 vs. 18.8%, P=0.001)with HES

-

Perner et al. (6s)11 rct severe sepsis 804 HES 130/0.42 ringer acetate 1 - Increased RRT need with HES(rr 1.35, 95% ci 1.01-1.80,

P=0.04)

Increased 90-day mortality with HES(rr 1.17, 95% ci 1.01-1.36, P=0.03)

Myburgh et al. (CHEST)12 rct General ICU 7000 HES 130/0.4 0.9% Nacl 1.2 More blood products in the HES group during the first 4 days

Increased RRT need with HES(7 vs. 5.8%, P=0.04)

No significant difference in 90-day mortality

Bayer et al.15 sequential observational study

severe sepsis 1046 4% gelatin crystalloids 1.1 More rBc and FFP transfusions in the gelatin group

Higher AKI rate(or 1.85, 95% ci 1.31-2.62)

andrrt need

(35.5 vs. 27.8%, P=0.033)with gelatin

No difference in mortality

thomas-rueddel et al.16 Meta-analysis critically ill, trauma, emergency, or elective surgery

patients

3275 gelatins crystalloids or albumin

- a trend toward increased exposure to allogeneic transfusions in patients

receiving gelatins

No difference in the risk of aKi a trend towards increased mortality in patients receiving gelatins


cardiac surgery 6478 4% gelatin crystalloids 1.1 More platelet concentrates in the gelatin group

Higher RRT need(or 2.75, 95% ci 1.84-4.16,

P<0.001)with gelatin

increased in-hospital mortality in the gelatin period

(or 1.72, 95% ci 1.15-2.58, P=0.008)*

Statistically significant differences are reported in italics. RCT: randomized controlled trials; ICU: intensive care unit; AKI: acute kidney injury; RRT: renal replacement therapy; RBC: red blood cells; FFP: fresh frozen plasma; RR: relative risk; CI: confidence interval; OR: odds ratio. *Multiple logistic regression analysis.

THE RISK OF INFUSING GELATIN PisaNo


hospitals, but with many more restrictions than indications. according to the eMa reso-lution,23 followed in italy by an aiFa notice,23 the use of HES solutions should be now only limited to “the treatment of hypovolemia due to acute blood loss when crystalloids alone are not considered sufficient”, while it is contrain-dicated in sepsis, burns, critically ill patients, patients with renal impairment or on rrt, and patients with severe coagulopathy. in the same period, also in USA the Food and Drug Ad-ministration (FDA) released a “boxed warn-ing” with similar restrictions.25

several subsequent meta-analyses con-firmed an increased risk of AKI,26, 27 rrt,27-31 red blood cell (rBc) transfusions,28, 29 and mortality 26, 28, 32 with HES administration, confirming that in ICU patients, especially those with sepsis, risks likely outweigh ben-efits also with “modern” HES solutions.

The use of synthetic colloids for fluid resus-

[CI] 1.01-1.36; P=0.03). The CHEST Trial, in which 7000 ICU patients (mostly surgical or septic) were randomized to receive the same volume of either 6% HES 130/0.4 or 0.9% sodium chloride (NaCl) for fluid resuscitation for 90 days after randomization or until ICU discharge or death, is the largest blinded rct on fluid therapy performed so far.12 although it failed to show any difference in mortal-ity (most likely because the amounts of toxic fluids were smaller than in previous studies), its key finding was a significantly increased RRT rate in the HES group as compared with the Nacl group (7 vs. 5.8%, P=0.04). More-over, post-hoc analyses showed a significant increase in serum creatinine levels during the first week and a significantly reduced overall urine output in the HES group as compared with the Nacl group (P=0.004 and P=0.003, respectively).

After a few months, HES was back in our

Table I.—�Main clinical investigations reporting major adverse effects of either hydroxyethyl starches (HES) or gelatins. Crystalloid/colloid ratios are also indicated when available.

study type of study Population N. colloid control crystalloid/ colloid ratio

adverse outcomes

transfusions renal Mortality

Brunkhorst et al. (ViseP)13 rct severe sepsis 537 HES 200/0.5 ringer lactate 1.3 - Higher AKI rate (34.9 vs. 22.8%, P=0.002)

andrrt need

(31 vs. 18.8%, P=0.001)with HES

-

Perner et al. (6s)11 rct severe sepsis 804 HES 130/0.42 ringer acetate 1 - Increased RRT need with HES(rr 1.35, 95% ci 1.01-1.80,

P=0.04)

Increased 90-day mortality with HES(rr 1.17, 95% ci 1.01-1.36, P=0.03)

Myburgh et al. (CHEST)12 rct General ICU 7000 HES 130/0.4 0.9% Nacl 1.2 More blood products in the HES group during the first 4 days

Increased RRT need with HES(7 vs. 5.8%, P=0.04)

No significant difference in 90-day mortality


severe sepsis 1046 4% gelatin crystalloids 1.1 More rBc and FFP transfusions in the gelatin group

Higher AKI rate(or 1.85, 95% ci 1.31-2.62)

andrrt need

(35.5 vs. 27.8%, P=0.033)with gelatin

No difference in mortality

thomas-rueddel et al.16 Meta-analysis critically ill, trauma, emergency, or elective surgery

patients

3275 gelatins crystalloids or albumin

- a trend toward increased exposure to allogeneic transfusions in patients

receiving gelatins

No difference in the risk of aKi a trend towards increased mortality in patients receiving gelatins


cardiac surgery 6478 4% gelatin crystalloids 1.1 More platelet concentrates in the gelatin group

Higher RRT need(or 2.75, 95% ci 1.84-4.16,

P<0.001)with gelatin

increased in-hospital mortality in the gelatin period

(or 1.72, 95% ci 1.15-2.58, P=0.008)*

Statistically significant differences are reported in italics. RCT: randomized controlled trials; ICU: intensive care unit; AKI: acute kidney injury; RRT: renal replacement therapy; RBC: red blood cells; FFP: fresh frozen plasma; RR: relative risk; CI: confidence interval; OR: odds ratio. *Multiple logistic regression analysis.



speed of shock reversal between patients with severe sepsis receiving either colloids or crys-talloids.15

Ultimately, if one of the reasons to use synthetic colloids is to minimize the risk of renal impairment (and mortality) which may be associated to fluid overload and interstitial edema,1-4, 37 it should be remembered that no large rct comparing colloids with crystal-loids found an increased risk of aKi or mor-tality with crystalloids.37 conversely, as men-tioned, most of them showed a higher rate of aKi and/or an increased risk of rrt need in patients receiving HES. After so many years of use, neither the efficacy nor the safety - as discussed in the following sections - of syn-thetic colloids have been proven,38 although some authors suggested that this may be due to methodological issues 39, 40 (maybe a clear symptom of how clinicians are fond of these drugs).

Gelatins impair coagulation at least as much as HES

gelatins are polypeptide molecules deriv-ing from bovine collagen which are available on the market (not in USA) as urea-linked gelatin, succinylated gelatin (modified fluid gelatin, MFg), and oxypolygelatin.37 In USA, the FDa withdrew the authorization for the clinical use of gelatin solutions as far back as in 1978 due to reduced coagulation and pro-longed bleeding time.40 in fact, in addition to dilutional coagulopathy to which all kind of fluids may contribute, all colloids can specifi-cally affect hemostasis due to an impairment of both coagulation (decreased von Willebrand and Viii factors, reduced thrombus formation, impaired interaction between thrombin and fi-brinogen and between factor XIII and fibrin, and reduced clot resistance to fibrinolysis) and platelet function.17 When compared with HES 130/0.4, gelatins have been found to reduce the strength of the fibrin clot to a lesser extent. However, the final effect is likely similar due to their slightly lower efficacy as volume-ex-panders. they also cause a more pronounced reduction in von Willebrand and Viii factors.42

citation in ICU patients was already progres-sively decreasing in some countries, such as australia and New Zealand, primarily due to a decreased use of gelatins and an increased use of buffered crystalloid solutions.33 However, unlike in USA and Australia,18 HES was wide-ly used in europe and, in particular, in italy, where many physicians believed that synthetic colloids have a much greater volume-expan-sion effect than crystalloids, and are thus indis-pensable when volume expansion is urgently needed. according to this rather ingrained be-lief (which is probably a misconception), and practically deprived of HES, European anes-thesiologists and intensivists might be tempted to rely once again on the “old” gelatins. They should not.

Colloids are better: reality or misconception?

Due to their longer persistence in the in-travascular compartment, colloids are tradi-tionally believed to be more effective than crystalloids as plasma volume expanders. ac-cordingly, they theoretically allow clinicians to achieve certain hemodynamic goals with a lower amount of fluids, thus potentially mini-mize the risk of volume overload and tissue edema. In most RCTs of fluid therapy, how-ever, the total volume of fluids administered in the crystalloid groups was only slightly higher than in the colloid groups, and the colloid:crystalloid ratio was never higher than 1:1.5,12, 13, 34-36 very far from the 1:4 ratio typi-cally taught in medical schools (see table i). The total amount of fluids administered was even not different between groups in the “6S” trial.11 similar results were found by Bayer et al. in two sequential analysis observational studies involving 6478 patients undergoing cardiac surgery 14 and 1046 patients with se-vere sepsis,15 respectively. Both investigations reported a colloid:crystalloid ratio of 1:1.4 for HES and, remarkably, of 1:1.1 for gelatin, suggesting that the supposed advantage of col-loids in terms of fluid balance is very modest, and even less prominent for gelatins. More-over, there was not significant difference in the



dergoing elective living-donor liver transplan-tation.45 Postoperative creatinine levels were significantly higher as compared to baseline in the gelatin group but not in the HES group. Moreover, postoperative estimated glomerular filtration rate was significantly lower and post-operative urine albumin:creatinine ratio was significantly higher, as compared to preopera-tive values, in the gelatin group. stage 1 aKi occurred in 5 patients in the gelatin group and in 2 patients in the HES group. Of course, no definitive conclusions can be drawn from this study.

thus, if renal adverse effects of gelatins are not worse than those of HES, they are most probably similar.14, 15, 37, 46 in the sequential analyses by Bayer et al., aKi rates and rrt needs were significantly higher in both sep-tic and cardiac surgery patients who received 4% gelatin (as well as in those receiving HES) compared with patients receiving crystalloids in the subsequent periods.14, 15

Maybe, gelatins are less nephrotoxic only when compared to the “older” HES solutions, as suggested by a multicenter rct, including 129 patients with severe sepsis, which found a lower rate of acute renal failure and oligu-ria, and a lower peak of serum creatinine, in patients receiving 3% MFg as compared with HES 200/0.6-0.66.47

Gelatins: another toxic and ineffective agent?

Bayer and colleagues found significantly increased in-hospital mortality among cardiac surgery patients who received 4% gelatin as compared with those who received crystal-loids for perioperative fluid therapy.14 after adjustment for several prognostic factors such as age, hypertension, diabetes mellitus, cirrho-sis, severe sepsis or septic shock, New York Heart Association class IV, Simplified Acute Physiology score ii, prior myocardial infarc-tion, pulmonary hypertension, cardiopulmo-nary bypass and aortic cross-clamp time, se-rum creatinine, and administration of specific drugs (noradrenaline, adrenaline, diuretics, angiotensin-converting-enzyme inhibitors,

Moreover, as well as HES solutions, both MFg and urea-linked gelatin solutions seem to inhibit platelet aggregation.

adverse effects of gelatins on coagulation similar to those of HES solutions are also sug-gested by the above mentioned sequential stud-ies.14, 15 cardiac surgery patients treated with either HES or gelatins received significantly more platelet concentrates than patients treated with crystalloids, although the number of pa-tients who needed rBc transfusions was not different among groups.14 Most remarkably, septic patients in both synthetic colloid groups (HES and gelatin) received significantly more rBc and fresh frozen plasma transfusions as compared with the crystalloid group.15 a trend toward an increased exposure to allogeneic transfusion in patients receiving gelatins as compared with patients receiving either crys-talloids or albumin was also found in a meta-analysis of 7 rcts.16 However, as pointed out by the authors, there was high heterogeneity among the included rcts, and none of them was adequately powered to assess the rate of relevant outcomes.

Gelatins might be even more nephrotoxic than HES

synthetic colloids may harm the kidney mainly due to osmotic nephrosis, consisting in cellular vacuolization and swelling in the proximal tubules, probably following pino-cytosis, and storage into vacuoles, of filtered colloid molecules.37 Unfortunately, the poten-tial renal toxicity of gelatins has never been studied in adequate rcts (probably because it is much less used than “modern” HES solu-tions among synthetic colloids). However, two animal studies found more severe renal injury with 4% gelatin as compared with 6% HES 130/0.4-0.42 in experimental models of sep-sis and shock.43,44 in particular, simon et al.44 found a significant increase in osmotic nephro-sis-like renal lesions with gelatin as compared with HES in a swine model of shock. These findings are consistent with those of a recent small rct comparing the renal effects of 4% gelatin and 6% HES 130/0.4 in 36 patients un-



Key messages

— according to randomized evidence of increased mortality and renal replace-ment therapy needs, the use of HES solu-tions has been recently highly restricted by the european Medicines agency with, in practice, many more contraindications than indications.

— since synthetic colloids are still be-lieved to be much more effective as vol-ume expanders than crystalloids (but this is probably largely overrated), physicians might be tempted to return to the “old” gelatins in order to replace HES.

— gelatins impairs coagulation at least as much as HES, and may be even more nephrotoxic than HES.

— as compared with crystalloids, gela-tins do not improve outcomes of critically ill patients (on the contrary, they might even increase mortality) and are much more ex-pensive. accordingly, their use should be discouraged.

References

1. Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid resuscitation in septic shock: a positive fluid bal-ance and elevated central venous pressure are associated with increased mortality. Crit Care Med 2011;39:259-65.

2. de Almeida JP, Palomba H, Galas FR, Fukushima JT, Duarte Fa, Nagaoka D, et al. Positive fluid balance is associated with reduced survival in critically ill patients with cancer. Acta Anaesthesiol Scand 2012;56:712-7.

3. lee J, de louw e, Niemi M, Nelson r, Mark rg, celi la, et al. Association between fluid balance and survival in critically ill patients. J Intern Med 2014;277:468-77.

4. Payen D, de Pont ac, sakr Y, spies c, reinhart K, Vin-cent Jl. sepsis occurrence in acutely ill Patients (soaP) Investigators: A positive fluid balance is associated with a worse outcome in patients with acute renal failure. crit Care 2008;12:R74.

5. Bouchard J, Soroko SB, Chertow GM, Himmelfarb J, ikizler ta, Paganini eP, Mehta rl. Program to improve care in acute renal Disease (PicarD) study group: Fluid accumulation, survival and recovery of kidney function in critically ill patients with acute kidney injury. Kidney Int 2009;76:422-7.

6. Fülöp t, Pathak MB, schmidt DW, lengvárszky Z, Jun-cos JP, lebrun cJ, et al. Volume-related weight gain and subsequent mortality in acute renal failure patients treat-ed with continuous renal replacement therapy. asaio J 2010;56:333-7.

7. grams Me, estrella MM, coresh J, Brower rg, liu KD. Fluid balance, diuretic use, and mortality in acute kidney injury. Clin J Am Soc Nephrol 2011;6:966-73.

non-steroidal anti-inflammatory drugs, some antimicrobials, iodinated contrast media), the odds ratios for in-hospital mortality with 4% gelatin vs. crystalloids were 1.72 (95% ci 1.15-2.58, P=0.008) and 1.40 (95% ci 1.07-1.84, P=0.016) on multiple logistic regression and propensity score stratification, respectively.

a trend toward increased mortality in pa-tients receiving gelatins compared with pa-tients receiving either crystalloids or albu-min was also found in a meta-analysis of 10 rcts.16 other investigations found similar mortality rates between fluid therapy with either gelatin or crystalloids in both patients with severe sepsis 15 and critically ill pa-tients.48 in particular, Perel et al. analyzed 11 rcts (including 506 patients, overall) and found no statistically significant difference in cumulative mortality between patients receiv-ing modified gelatin or crystalloids (risk ratio 0.91, 95% ci 0.49-1.72).48 in addition to pos-sibly increasing mortality, gelatins are much more expensive than crystalloids.48 according to this consideration alone, their use appears to be unjustified. Moreover, gelatins probably carry a higher risk of anaphylactic reactions as compared to other colloids.37 this issue be-came evident soon after their introduction into clinical practice.49, 50 there are many reports in literature of perioperative (or periprocedural) severe anaphylactic reactions due to the ad-ministration of intravenous gelatins,51-53 with fatal events reported also very recently.53 an-other reason (if it is really needed) to avoid them.

Conclusions

In conclusion, the era of artificial colloids is coming to an end. it is likely that history of medicine will reflect in amazement on how doctors decided to give extracts of either corn or potatoes (starches) or extracts of horses’ hooves (gelatins) intravenously to patients, believing that such interventions would be beneficial. Modern evidence-based medicine is clear on this issue: artificial colloids are harmful and cost much more than crystalloids. there is simply no reason to prescribe them.



24. aiFa. [important information notice on the restriction of use of HES (hydroxyethyl starch-containing medicines)]; 2013 [internet]. available from: http://www.agenziafar-maco.gov.it/sites/default/files/IT_DHPC_HES_common.pdf [cited 2015, Dec 8].

25. Food and Drug administration. FDa safety communica-tion: boxed warning on increased mortality and severe renal injury, and additional warning on risk of bleeding, for use of hydroxyethyl starch solutions in some settings; 2013 [internet]. available from: http://www.fda.gov/Bio-logicsBloodVaccines/safetyavailability/ucm358271.htm [cited 2015, Dec 8].

26. Zarychanski R, Abou-Setta AM, Turgeon AF, Houston Bl, Mcintyre l, Marshall Jc, et al. association of hy-droxyethyl starch administration with mortality and acute kidney injury in critically ill patients requiring volume resuscitation: a systematic review and meta-analysis. JAMA 2013;309:678-88.

27. Mutter TC, Ruth CA, Dart AB. Hydroxyethyl starch (HES) versus other fluid therapies: effects on kidney func-tion. Cochrane Database Syst Rev 2013;7:CD007594.

28. Patel A, Waheed U, Brett SJ. Randomised trials of 6% tet-rastarch (hydroxyethyl starch 130/0.4 or 0.42) for severe sepsis reporting mortality: systematic review and meta-analysis. Intensive Care Med 2013;39:811-22.

29. Haase N, Perner A, Hennings LI, Siegemund M, Lau-ridsen B, Wetterslev M, et al. Hydroxyethyl starch 130/0.38–0.45 versus crystalloid or albumin in patients with sepsis: systematic review with meta-analysis and trial sequential analysis. BMJ 2013;346:f839.

30. gattas DJ, Dan a, Myburgh J, Billot l, lo s, Finfer s, CHEST Management Committee. Fluid resuscitation with 6 % hydroxyethyl starch (130/0.4 and 130/0.42) in acutely ill patients: systematic review of effects on mor-tality and treatment with renal replacement therapy. in-tensive Care Med 2013;39:558-68.

31. rochwerg B, alhazzani W, gibson a, ribic cM, sindi A, Heels-Ansdell D, et al. FISSH Group (Fluids in Sepsis and septic shock): Fluid type and the use of renal re-placement therapy in sepsis: a systematic review and net-work meta-analysis. Intensive Care Med 2015;41:1561-71.

32. Rochwerg B, Alhazzani W, Sindi A, Heels-Ansdell D, thabane l, Fox-robichaud a, et al. Fluids in sepsis and septic shock group: Fluid resuscitation in sepsis: a sys-tematic review and network meta-analysis. ann intern Med 2014;161:347-55.

33. Hammond NE, Taylor C, Saxena M, Liu B, Finfer S, glass P, et al. Resuscitation fluid use in Australian and New Zealand Intensive Care Units between 2007 and 2013. Intensive Care Med 2015;41:1611-9.

34. guidet B, Martinet o, Boulain t, Philippart F, Poussel JF, Maizel J, et al. Assessment of hemodynamic efficacy and safety of 6% hydroxyethylstarch 130/0.4 vs. 0.9% NaCl fluid replacement in patients with severe sepsis: the CRYSTMAS study. Crit Care 2012;16:R94.

35. James MF, Michell Wl, Joubert ia, Nicol aJ, Navsaria PH, Gillespie RS. Resuscitation with hydroxyethyl starch improves renal function and lactate clearance in penetrat-ing trauma in a randomized controlled study: the First trial (Fluids in resuscitation of severe trauma). Br J Anaesth 2011;107:693-702.

36. Finfer s, Bellomo r, Boyce N, French J, Myburgh J, Norton r. saFe study investigators: a comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004;350:2247-56.

37. Mårtensson J, Bellomo R. Are all fluids bad for the kid-ney? Curr Opin Crit Care 2015;21:292-301.

38. canet J, sabaté s, Mazo V. colloids administration in surgery: safety and efficacy still unproven. Minerva An-estesiol 2014;80:858-9.

8. reNal replacement therapy study investigators, Bel-lomo r, cass a, cole l, Finfer s, gallagher M, lee J, et al. An observational study fluid balance and patient outcomes in the randomized evaluation of Normal vs. augmented level of replacement therapy trial. crit Care Med 2012;40:1753-60.

9. Vaara st, Korhonen aM, Kaukonen KM, Nisula s, ink-inen O, Hoppu S, et al. Fluid overload is associated with an increased risk for 90-day mortality in critically ill pa-tients with renal replacement therapy: data from the pro-spective FINNAKI study. Crit Care 2012;16:R197.

10. teixeira c, garzotto F, Piccinni P, Brienza N, iannuzzi M, gramaticopolo s, et al. Fluid balance and urine vol-ume are independent predictors of mortality in acute kid-ney injury. Crit Care 2013;17:R14.

11. Perner A, Haase N, Guttormsen AB, Tenhunen J, Kle-menzson g, Åneman a, et al. Hydroxyethyl starch 130/0.42 versus ringer’s acetate in severe sepsis. N engl J Med 2012;367:124-34.

12. Myburgh Ja, Finfer s, Bellomo r, Billot l, cass a, gat-tas D, et al. Hydroxyethyl starch or saline for fluid resus-citation in intensive care. N Engl J Med 2012;367:1901-11.

13. Brunkhorst FM, Engel C, Bloos F, Meier-Hellmann A, ragaller M, Weiler N, et al. intensive insulin therapy and pentastarch resuscitation in severe sepsis. N engl J Med 2008;358:125-39.

14. Bayer o, schwarzkopf D, Doenst t, cook D, Kabisch B, schelenz c, et al. Perioperative fluid therapy with tetras-tarch and gelatin in cardiac surgery-a prospective sequen-tial analysis*. Crit Care Med 2013;41:2532-42.

15. Bayer o, reinhart K, Kohl M, Kabisch B, Marshall J, sakr Y, et al. Effects of fluid resuscitation with synthetic colloids or crystalloids alone on shock reversal, fluid balance, and patient outcomes in patients with severe sepsis: a prospective sequential analysis. crit care Med 2012;40:2543-51.

16. thomas-rueddel Do, Vlasakov V, reinhart K, Jaeschke R, Rueddel H, Hutagalung R, et al. safety of gelatin for volume resuscitation-a systematic review and meta-anal-ysis. Intensive Care Med 2012;38:1134-42.

17. Kozek-langenecker sa: Fluids and coagulation. curr Opin Crit Care 2015;21:285-91.

18. Finfer s, liu B, taylor c, Bellomo r, Billot l, cook D, et al. Resuscitation fluid use in critically ill adults: an international cross-sectional study in 391 intensive care units. Crit Care 2010;14:R185.

19. european Medicines agency. Prac recommends sus-pending marketing authorizations for infusion solu-tions containing hydroxyethyl starch; 2013 [Internet]. available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/news_and_events/news/2013/06/news_detail_001814.jsp&mid=WC0b01ac058004d5c1 [cited 2015, Dec 8].

20. Wise J. Boldt: the great pretender. BMJ 2013;346:f1738.21. Wiedermann cJ, Wiedermann W. Beautiful small: Mis-

leading large randomized controlled trials? the example of colloids for volume resuscitation. J anaesthesiol clin Pharmacol 2015;31:394-400.

22. Wiedermann CJ, Dunzendorfer S, Gaioni LU, Zaraca F, Joannidis M. Hyperoncotic colloids and acute kidney injury: a meta-analysis of randomized trials. crit care 2010;14:R191.

23. European Medicines Agency. Hydroxyethyl-starch so-lutions (HES) no longer to be used in patients with sepsis or burn injuries or in critically ill patients; 2013 [internet]. available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/referrals/Hydroxyethyl_starch-containing_medicines/human_re-ferral_prac_000029.jsp&mid=WC0b01ac05805c516f [cited 2015, Dec 8].



46. Davidson I. Renal impact of fluid management with colloids: a comparative review. eur J anaesthesiol 2006;23:721-38.

47. Schortgen F, Lacherade JC, Bruneel F, Cattaneo I, Hem-ery F, lemaire F, et al. effects of hydroxyethylstarch and gelatin on renal function in severe sepsis: a multicentre randomised study. Lancet 2001;357:911-6.

48. Perel P, roberts i, Ker K. colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Da-tabase Syst Rev 2013 Feb 28;2:CD000567

49. isbister J, Fisher M. adverse effects of plasma volume expanders. Anaesth Intensive Care 1980;8:145-51

50. ring J, Messmer K. incidence and severity of anaphy-lactoid reactions to colloid volume substitutes. lancet 1977;1:466-9.

51. apostolou e, Deckert K, Puy r, sandrini a, de leon MP, Douglass Ja, et al. Anaphylaxis to Gelofusine confirmed by in vitro basophil activation test: a case series. anaes-thesia 2006;61:264-8.

52. Dubrey sW, Dahdal g, grocott-Mason r. severe anaph-ylaxis to gelofusine during a transthoracic echo bubble study. Eur J Echocardiogr 2008;9:303.

53. Ventura spagnolo e, calapai g, Minciullo Pl, Mannucci c, asmundo a, gangemi s. lethal anaphylactic reac-lethal anaphylactic reac-tion to intravenous gelatin in the course of surgery. am J Ther 2015; Apr 17 [Epub ahead of print].

39. orbegozo cortes D, santacruz c, Donadello K, Nobile l, taccone Fs. Colloids for fluid resuscitation: what is their role in patients with shock? Minerva anestesiol 2014;80:963-9.

40. aldecoa c, Kozek-langenecker s, rico-Feijoo J. col-loids in surgery: bad drugs, bad protocol, or bad data analysis? Minerva Anestesiol 2014;80:856-7.

41. Kozek-Langenecker SA. Influence of fluid therapy on the haemostatic system of intensive care patients. Best Pract Res Clin Anaesthesiol 2009;23:225-36.

42. list of Drug Products that have been withdrawn or re-moved from the market for the reasons of safety or effec-tiveness; [Internet]. Available from: http://www.fda.gov/ohrms/dockets/98fr/da1008co.pdf [cited 2015, Dec 8].

43. schick Ma, isbary tJ, schlegel N, Brugger J, Waschke J, Muellenbach r, et al. the impact of crystalloid and colloid infusion on the kidney in rodent sepsis. intensive Care Med 2010;36:541-8.

44. Simon TP, Schuerholz T, Hüter L, Sasse M, Heyder F, Pfister W, Marx G. Impairment of renal function using hyperoncotic colloids in a two hit model of shock: a pro-spective randomized study. Crit Care 2012;16:R16.

45. Demir A, Aydınlı B, Toprak HI, Karadeniz Ü, Yılmaz FM, Züngün c, et al. impact of 6% starch 130/0.4 and 4% gelatin infusion on Kidney Function in living-Donor Liver Transplantation. Transplant Proc 2015;47:1883-9.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: April 5, 2016. - Manuscript accepted: March 31, 2016. - Manuscript revised: March 1, 2016. - Manu-script received: December 10, 2015.


monary interactions in mechanically ventilated patients in sinus rhythm, previously described as the functional hemodynamic monitoring.6 these dynamic indices constitute simple, widely available tools for bedside use with the high-risk surgical patients in the intensive care unit (icU) or the operating theater. the simplest, most popular tool is pulse pressure variation (PPV) measurement, which was ini-tially described in septic, critically ill patients.7 thereafter, the stroke volume variation (sVV) has been described as accurately predict fluid responsiveness.8 recently, some promising

Fluid loading remains a major problem in the routine care of critically ill patients

and high-risk surgical patients. Most of the ap-proaches currently used are empirical and not evidence-based.1 this is not the case of the mini-fluid challenge,2 the passive leg-raising test (Plr) 3 and the end-expiratory occlusion maneuver,4 which are reportedly good predic-tors of fluid responsiveness when cardiac out-put monitoring is used, although the last tech-nique is not always available at the bedside.1, 5 Another way of predicting fluid responsiveness involves dynamic indices based on cardiopul-


a dynamic view of dynamic indicesMarc-olivier FiscHer 1, 2 *, Pierre-grégoire gUiNot 3, 4,

Matthieu Biais 5, 6, Yazine MaHJoUB 3, 4, Jihad Mallat 7, emmanuel lorNe 3, 4 on behalf of the French Hemodynamic team (FHt)

1Pôle réanimations anesthésie saMU/sMUr, cHU de caen, caen, France; 2ea 4650, Université caen Normandie, caen, France; 3Department of anesthesiology and critical care, amiens University Medical center, amiens, France; 4iNserM eri12, Jules Verne University of Picardy, amiens, France; 5Department of anesthesiology and critical care Medicine iii, Bordeaux University Medical center, Bordeaux, France; 6“adaptation cardiovasculaire à l’ischémie” research Unit, iNserM U1034, University of Bordeaux, Pessac, France; 7Department of anesthesiology and critical care Medicine, centre Hospitalier du Dr schaffner, lens, France*corresponding author: Marc-olivier Fischer, Pôle réanimations anesthésie saMU/sMUr, cHU de caen, avenue de la côte de Nacre, CS 30001, F-14000 Caen, France. E-mail: [email protected]


lavoro: 11244-Mastitolo breve: DYNaMic iNDicesprimo autore: FiscHerpagine: 1115-21citazione: Minerva anestesiol 2016;82:1115-21

a B s t r a c tDynamic indices (based on cardiopulmonary interactions in mechanically ventilated patients in sinus rhythm) have been developed as simple tools for predicting fluid responsiveness in the absence of cardiac output monitoring. Although the earliest dynamic indices relied on the invasive measurement of pulse pressure variations or stroke volume variations, the most recently developed indices are based on non-invasive photoplethysmography. However, a number of confounding factors have been found which decrease the clinical value of these indices. the present experts’ opinion explains why changes in dynamic indices during hemodynamic maneuvers might be an interesting alternative to using them accurately at bedside.(Cite this article as: Fischer Mo, guinot Pg, Biais M, Mahjoub Y, Mallat J, lorne e; French Hemodynamic team (FHt). a dynamic view of dynamic indices. Minerva anestesiol 2016;82:1115-21)Key words: cardiac output - Hemodynamics - Physiologic monitoring - Plethysmography - Blood pressure - stroke volume.


FiscHer DYNaMic iNDices


manufacturers have developed hemodynamic monitors that include PPV in standard bed-side monitors (intelliVue, Phillips Medical systems, andover, Ma, Usa; carescape, ge Healthcare, Wi, Usa), and in bedside designed hemodynamic monitors (PulsioFlex and Picco2, both by Pulsion Medical, Mu-nich, germany; Floctrac and eV1000, both by edwards lifescience, irvine, ca, Usa; lidco rapid/Unity, liDco ltd., cambridge, england, UK; Mostcare, Vytech Health, lab-oratoires Pharmaceutiques Vygon, ecouen, France).

Non-invasive PPV

totally non-invasive measurement systems (based on photoplethysmography) are now of-fered by some manufacturers (e.g. the clear-sight system by edwards lifesciences, the former Nexfin device [BMYe, Amsterdam, the Netherlands], and the CNAP system [CNS systems Medizintechnik ag, graz, austria]). the PPV is averaged over different time mov-ing periods. these automated, plug-and-play devices use a finger sensor connected to a dedicated monitor, which meets the simplified criteria needed at the bedside.18

Stroke volume variation

in addition to PPV, the sVV was introduced as a dynamic preload indicator that is calculat-ed from percentage changes in stroke volume (sV) during ventilatory cycle. Many manufac-turers have developed hemodynamic monitors that include sVV (esophageal Doppler moni-toring in addition to the devices previously de-scribed for PPV). calculation of sVV is based on the percentage of change between the maxi-mal and minimal sVs divided by the average of the minimum and maximum over a floating period,8 or on the difference between maxi-mal and minimal sVs divided by the average of the minimum and maximum sV over one respiratory cycle.19 sVV was demonstrated to be a good indicator of fluid responsiveness in operating room and icU.19, 20

non-invasive approaches (using continuous, non-invasive arterial pressure monitoring9, 10 or the changes in the plethysmographic wave-form in mechanically ventilated patients 11, 12) have been described. Dynamic indices over static variables are recommended for predict-ing fluid responsiveness, when applicable.13 However, confounding factors may decrease the clinical value of these dynamic indices.14 a potentially valuable alternative is the dy-namic use of dynamic indices, i.e. the use of rapid changes in these indices (PPV, sVV, and changes in the plethysmographic waveform in mechanically ventilated patients) during he-modynamic maneuvers to accurately predict fluid responsiveness without interpretation bias.15

Hence, the present experts’ opinion seeks to 1) describe the currently available tools and their limitations, 2) develop the “dynamic use of dynamic indices” concept, and 3) explore the perspectives for these techniques.

The currently available dynamic indices

Invasive PPV

The PPV was firstly described using the fol-lowing formula:

PPV = (PPmax – PPmin)/[(PPmax + PPmin) × 0.5]

where PPmax and PPmin represent the maxi-mal and the minimal amplitude of the arterial pulse pressure waveform over one respiratory cycle, respectively.7 the PPV was initially calculated manually over a single respiratory cycle. Using a 60-second moving window, the PPV is now assessed accurately and auto-matically based on the arterial blood pressure alone, which eliminates the need of simulta-neous recording of airway pressure.16 From a physiological point of view, as the pulse pres-sure was correlated with stroke volume, PPV could follow the volume changes in arteries. From the free by aboy,16 several different algorithms of PPV calculation were devel-oped by some manufacturers using different time moving windows and signal interpreta-tion such as extrasystole exclusion.17 Many

DYNaMic iNDices FiscHer


change a dynamic index’s threshold value. right ventricular dysfunction 25 and intra-ab-dominal hypertension26 may increase the PPV threshold. in contrast, a low tidal volume (<8 ml/kg),27 low pulmonary compliance (<30 cm-H2o/ml) 4 and a low heart rate/respiratory rate ratio (<3.6) 28 may decrease the PPV threshold (Figure 1). thirdly, changes in the threshold value may explain a wide inconclusive class of response. the proportion of patients in this “gray zone” is reportedly about 25% in the op-erating theater (with PPV values between 9% and 13%) 29 and 62% in the icU (with PPV values between 4 and 17%).30 Fourthly, the PPV is difficult to interpret at the bedside. In a recent survey of French anesthesiologists and intensivists (based on clinical vignettes), only 60% of the responders were aware of the three prerequisites for PPV validity, and none was able to interpret the PPV correctly with regard to all the confounding factors.15 lastly, dynamic indices have rarely been evaluated in large, randomized studies, although encourag-ing results have been reported in studies with a small number of patients and widely differ-ing protocols and optimization targets (mean arterial pressure, cardiac index, stroke vol-ume, central venous pressure, central venous oxygen saturation, etc.). in view of this great heterogeneity, the conclusions of a recent me-ta-analysis may be questionable.31

Although dynamic indices were firstly de-scribed as a simple tool for predicting fluid

Non-invasive SVV

totally non-invasive measurement systems (based on photoplethysmography) are now of-fered by some manufacturers (e.g. the clear-Sight system, the former Nexfin device, and the cNaP system).

The delta-POP and the Plethysmographic Variability Index (PVI)

Using usual pulse oximetry sensor, the anal-ysis of changes in the plethysmographic wave-form in mechanically ventilated patients has been proposed as a non-invasive alternative to PPV. The delta-POP was firstly described 11 us-ing the following formula:

Delta-PoP = (PoPmax – PoPmin)/ [(POPmax + PoPmin) × 0.5]

where PoPmax and PoPmin represent the maxi-mal and the minimal amplitude of the ple-thysmographic waveform over one respira-tory cycle, respectively.11 the delta-PoP was manually calculated, in contrast with the PVi which is continuously displays for a 2-to-4-minute moving window.12 the PVi needs a dedicated monitor (e.g. the radical 7 by Ma-simo corporation, irvine, ca, Usa). a recent meta-analysis showed that delta-PoP and PVi are equally effective for predicting fluid re-sponsiveness in mechanically ventilated adult patients in sinus rhythm.21

Limitations of dynamic indices

Limitations that apply to all dynamic indices

over the last ten years, reports of various limitations have fueled the continuing debate on the clinical value of dynamic indices.22, 23

Firstly, the validity of a dynamic index de-pends on three mandatory prerequisites: con-tinuous arterial pressure monitoring, a regular sinus rhythm, and mechanical ventilation (i.e. in the absence of spontaneous breathing).7 re-cent surveys showed that 18% to 42% of criti-cally ill patients met all three criteria.14, 24 sec-ondly, a number of confounding factors may

Figure 1.—changes in the threshold value of dynamic in-dices.

Low %dal volume Low pulmonary compliance Low heart rate/respiratory rate ra%o

Right ventricular dysfunc%on Intra-‐abdominal hypertension

Cut-‐off value in valida%on studies 13 %

>13 %

<13 %

Changes in the threshold value of dynamic indices



PVi in the icU is subject to debate.34-37 inter-estingly, it was recently reported that a fore-head sensor could be more accurate than a fin-ger sensor for predicting fluid responsiveness in critically ill patients with high vasomotor tone.38

The dynamic use of dynamic indices

Efficacy of fluid challenge

in contrast to changes in arterial pres-sure components, changes in an invasively measured PPV were able to detect a fluid-challenge-induced increase in cardiac output of more than 15% with good sensitivity and specificity, and with only a small proportion of patients in the “gray zone”.39 in view of the relationship between fluid loading and cardiac output, this approach may enable clinicians to proceed with or stop fluid loading accord-ingly; fractional fluid loading could be guided by changes in the PPV. of course, this strategy would need to be evaluated in dedicated stud-ies. encouraging results have been recently described for the PPV- and Plr-based titra-tion of fluid responsiveness in septic patients; guidance with preload dependence indices was associated with a lower daily fluid intake than a control technique based on central venous pressure guidance, and did not worsen the clinical outcome.40

Mini-fluid challenge

recently, a decrease in PPV or sVV during a mini-fluid challenge (infusion of 100 mL of 4% human serum albumin over 60 seconds) was found to accurately predict fluid respon-siveness in septic, mechanically ventilated pa-tients with a low tidal volume.41 Whereas the baseline PPV and sVV values were not pre-dictive of fluid responsiveness, the change in PPV or SVV induced by the mini-fluid chal-lenge yielded excellent sensitivity and speci-ficity values. The small proportion of patients (12%) in the “gray zone” further emphasized the clinical applicability of the “dynamic use of dynamic indices”. interestingly, the changes

responsiveness, their application requires a degree of caution. accordingly, sondergaard suggested a checklist before using PPV to guide fluid therapy.23

Specific limitations of non-invasive PPV and SVV measurements

Few studies have compared non-invasive PPV measurements with the invasive PPV measurements for predicting fluid responsive-ness, with controversial results.9, 32

Specific limitations of delta-POP and PVI

For example, calculation of the PVi is based on changes in the Perfusion index (Pi; the ratio between the alternating current and the direct current components), as follows:

PVi = (Pimax – Pimin)/Pimax

where Pimax and Pimin represent the maximal and the minimal amplitude of the plethysmo-graphic waveform over one respiratory cycle, respectively.12 this equation explains two limitations: 1) small changes in Pi can lead to large changes in PVi, and 2) the Pi depends on sympathetic vasomotor tone as well as the stroke volume. Precisely, the pulsatile ple-thysmographic and photo-plethysmographic amplitude is due to pulsatile changes of the arteriolar blood volume into the tissue under the beam. The blood volume pulsations (ΔV) are related to the systemic intravascular pulse pressure (ΔP) according to the relationship: ∆V=∆P×D, where D is the distensibility that is influenced by intravascular volume status and by sympathetic activity directed to the vessels. as demonstrated by colombo et al., the indices derived from the pulsatile photo-plethysmography are affected by a stimuli that activate the sympathetic branch of the autonomic nervous system.33 in this context, a (relative) hypovolemia represent probably only one stimulus among many others. in-deed, pain, hypothermia and vasopressor use can decrease the PVI’s ability to predict fluid responsiveness.34 these explanations could explain why the value of using a non-invasive



criminate responders and non-responders to vasopressor dose de-escalation with volume expansion.47 in sepsis patients receiving nor-epinephrine, the eadyn predict the decrease in arterial pressure following a norepinephrine dose reduction.45 Finally, eadyn might consti-tute an easy-to-use, functional tool for arterial-tone assessment and thus may help to identify patients likely to benefit from vasopressor treatment.

Right ventricular dysfunction

PPV could not accurately predict fluid re-sponsiveness in heart surgery patients and septic, critically ill patients with increased pul-monary artery pressure.48 Patients with right ventricular dysfunction (defined as a peak sys-tolic velocity of tricuspid annular motion <0.15 m/s, as assessed by tissue Doppler echocar-diography) may have a higher PPV threshold; if so, this would explain a large proportion of “false-positive” preload-dependent patients.25 However, the absence of a decrease or indeed an increase in dynamic indices after fluid load-ing (when the baseline PPV is high) should be considered as an indicator of right ventricular dysfunction.49 in this context, changes in dy-namic indices could usefully indicate the need for echocardiography.

Conclusions

according to recent guidelines, dynamic variables (rather than static variables) should be used to predict fluid responsiveness (when applicable), and fluid resuscitation should be guided by monitoring more than one he-modynamic variable. caution must be used when dynamic indices are used alone (in view of their many limitations). at the bedside, changes in dynamic indices (rather than abso-lute values) may be more predictive of fluid responsiveness and may thus avoid the ef-fects from confounding factors (i.e. the “gray zone”). large-scale validation studies are now mandatory for confirming this hypothesis and defining perspectives for the reliable clinical use of dynamic indices.

in PPV or sVV gave better results than chang-es in continuous cardiac output monitoring (a calibrated pulse contour analysis using the Picco2 device). Hence, a strategy based on changes in PPV or sVV would be less inva-sive than the assessments of fluid responsive-ness described above.

Passive leg raising test

although baseline PVi can detect cardiac output changes following Plr in spontane-ously breathing volunteers,42 there are no data on the ability of changes in dynamic indices during PLR to predict fluid responsiveness.

Future perspectives for dynamic indices

Combinations of dynamic indices

the combination of dynamic indices might increase the predictive value for fluid respon-siveness. in a comparative study of three PVi measurement sites, combining the Pi and PVi thresholds increased the specificity and the positive and negative predictive values for pre-dicting fluid responsiveness.43 Further studies are now required for the development of re-classification methods and progressive models (based on various hemodynamic variables and maneuvers) able of safely predicting fluid re-sponsiveness at the bedside.

Assessment of dynamic arterial elastance

encouraging results have been reported for use of the PPV/sVV ratio as a guide to dy-namic arterial elastance (eadyn) in operating theater or icU, in mechanically ventilated pa-tients or spontaneously breathing patients.44, 45 eadyn may constitute a functional approach to arterial tone assessment in the same way as preload responsiveness indices that are used to predict fluid responsiveness to a change in cardiac preload. eadyn was able to predict the hemodynamic response in mean arterial pres-sure (MAP) to fluid administration in hypoten-sive, preload-dependent patients.46 in surgical patients, eadyn was found to successfully dis-



using a finger cuff corresponds with intra-arterial meas-urement. Br J anaesth 2011;107:540-5.

10. stens J, oeben J, Van Dusseldorp aa, Boer c. Non-inva-sive measurements of pulse pressure variation and stroke volume variation in anesthetized patients using the Nex-fin blood pressure monitor. J Clin Monit Comput 2015;in press.

11. Desebbe o, cannesson M. Using ventilation-induced plethysmographic variations to optimize patient fluid sta-tus. curr opin anaesthesiol 2008;21:772-8.

12. cannesson M, slieker J, Desebbe o, Bauer c, chiari P, Hénaine r, et al. the ability of a novel algorithm for au-tomatic estimation of the respiratory variations in arterial pulse pressure to monitor fluid responsiveness in the op-erating room. anesth analg 2008;106:1195-2000.

13. cecconi M, De Backer D, antonelli M, Beale r, Bakker J, Hofer c, et al. consensus on circulatory shock and hemodynamic monitoring. task force of the european society of intensive care Medicine. intensive care Med 2014;40:1795-815.

14. Fischer Mo, Mahjoub Y, Boisselier c, tavernier B, Dupont H, leone M, et al. arterial pulse pressure vari-ation suitability in critical care: a French national survey. anaesth crit care Pain Med 2015;34:23-8.

15. Fischer Mo, Dechanet F, du cheyron D, gérard Jl, Ha-nouz Jl, Fellahi Jl. evaluation of the knowledge base of French intensivists and anaesthesiologists as concerns the interpretation of respiratory arterial pulse pressure varia-tion. anaesth crit care Pain Med 2015;34:29-34.

16. aboy M, McNames J, thong t, Philips cr, ellenby Ms, goldstein B. a novel algorithm to estimate the pulse pressure variation index deltaPP. ieee trans Biomed eng 2004;51:2198-203.

17. cannesson M, tran NP, cho M, Hatib F, Michard F. Predicting fluid responsiveness with stroke volume variation despite multiple extrasystoles. crit care Med 2012;40:193-8.

18. De Waal ee, Wappler F, Buhre WF. cardiac output moni-toring. curr opin anaesthesiol 2009;22:71-77.

19. guinot Pg, de Broca B, abou arab o, Diouf M, Badoux l, Bernard e, et al. ability of stroke volume variation measured by oesophageal Doppler monitoring to pre-dict fluid responsiveness during surgery. Br J Anaesth. 2013;110:28-33.

20. Hofer cK, senn a, Weibel l, Zollinger a. assessment of stroke volume variation for prediction of fluid respon-siveness using the modified FloTrac and PiCCOplus sys-tem. crit care 2008;12:r82.

21. sandroni c, cavallaro F, Marano c, Falcone c, De santis P, antonelli M. accuracy of plethysmographic indices as predictors of fluid responsiveness in mechanically venti-lated adults: a systematic review and meta-analysis. in-tensive care Med 2012;38:1429-37.

22. Michard F, chemla D, teboul Jl. applicability of pulse pressure variation: how many shades of grey? crit care 2015;19:144.

23. sondergaard s. Pavane for a pulse pressure variation de-funct. crit care 2013;17:327.

24. Benes J, Zatloukal J, Kletecka J, simanova a, Haid-ingerova l, Pradl r. respiratory induced dynamic varia-tions of stroke volume and its surrogates as predictors of fluid responsiveness: applicability in the early stages of specific critical states. J Clin Monit Comput 2014;28:225-31.

25. Mahjoub Y, Pila c, Friggeri a, Zogheib e, lobjoie e, tinturier F, et al. Assessing fluid responsiveness in criti-cally ill patients: False-positive pulse pressure variation is detected by Doppler echocardiographic evaluation of the right ventricle. crit care Med 2009;37:2570-5.

26. Mahjoub Y, touzeau J, airapetian N, lorne e, Hijazi M, Zogheib e, et al. the passive leg-raising maneuver

Key messages

— Dynamic indices were developed to predict fluid responsiveness in mechani-cally ventilated and sinus rhythm patients. However, caution must be used when dy-namic indices are used alone, because of their many confounding factors that may change their cut-off value.

— at the bedside, changes in dynamic indices (rather than absolute values) may be more predictive of fluid responsiveness or efficacy of fluid challenge, and may thus avoid the effects from confounding factors.

— large-scale validation studies are now mandatory for confirming this hypoth-esis and defining perspectives for the reli-able clinical use of dynamic indices.

References

1. cecconi M, Hofer c, teboul Jl, Petilla V, Wilkman e, Molnar Z, et al. Fluid challenges in intensive care: the FeNice study. a global inception cohort study. intensive care Med 2015;41:1529-37.

2. Muller l, toumi M, Bousquet PJ, riu-Poulenc B, louart g, candela D, et al. An increase in aortic blood flow after an infusion of 100 ml colloid over 1 minute can predict fluid responsiveness: the mini-fluid challenge study. An-esthesiology 2011;115:541-7.

3. Monnet X, rienzo M, osman D, anguel N, rich-ard c, Pinsky Mr, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med 2006;34:1402-7.

4. Monnet X, Bleibtreu a, Ferre a, Dres M, gharbi r, ri-chard c, et al. Passive leg-raising and end-expiratory oc-clusion tests perform better than pulse pressure variation in patients with low respiratory system compliance. crit care Med 2012;40:152-7.

5. ahmad t, Beilstein cM, aldecoa c, Moreno rP, Mol-nar Z, Novak-Jankovic V, et al. Variation in haemody-namic monitoring for major surgery in european nations: secondary analysis of the eusos dataset. Perioper Med 2015;4:8.

6. cavallaro F, sandroni c, antonelli M. Functional hemo-dynamic monitoring and dynamic indices of fluid respon-siveness. Minerva anestesiol 2008;74:123-35.

7. Michard F, Boussat s, chemla D, anguel N, Mercat a, lecarpentier Y, et al. relation between respiratory changes in arterial pulse pressure and fluid responsive-ness in septic patients with acute circulatory failure. am J respir crit care Med 2000;162:134-8.

8. Berkenstadt H1, Margalit N, Hadani M, Friedman Z, segal e, Villa Y, et al. stroke volume variation as a predictor of fluid responsiveness in patients undergoing brain surgery. anesth analg 2001;92:984-9.

9. lansdorp B, ouweneel D, de Keijzer a, van der Hoeven Jg, lemson J, Pickkers P. Non-invasive measurement of pulse pressure variation and systolic pressure variation



variability index measurement: a comparative pilot study in cardiac surgery patients. J cardiothorac Vasc anesth 2014;28:1510-5.

39. le Manach Y, Hofer cK, lehot JJ, Vallet B, goarin JP, tavernier B, et al. can changes in arterial pressure be used to detect changes in cardiac output during volume expansion in the perioperative period? anesthesiology 2012;117:1165-74.

40. richard Jc, Bayle F, Bourdin g, leray V, Debord s, Delannoy B, et al. Preload dependence indices to titrate volume expansion during septic shock: a randomized controlled trial. crit care 2015;19:5.

41. Mallat J, Meddour M, Durville e, lemyze M, Pepy F, temime J, et al. Decrease in pulse pressure and stroke volume variations after mini-fluid challenge accurately predicts fluid responsiveness. Br J Anaesth 2015;115:449-56.

42. Keller g, cassar e, Desebbe o, lehot JJ, cannesson M. ability of pleth variability index to detect hemodynamic changes induced by passive leg raising in spontaneously breathing volunteers. crit care 2008;12:r37.

43. Desgranges FP, Desebbe o, ghazouani a, gilbert K, Kel-ler g, chiari P, et al. Influence of the site of measurement on the ability of plethysmographic variability index to predict fluid responsiveness. Br J Anaesth 2011;107:329-35.

44. cecconi M, Monge garcía Mi, gracia romero M, Mel-linghoff J, caliandro F, grounds rM, et al. the use of pulse pressure variation and stroke volume variation in spontaneously breathing patients to assess dynamic arte-rial elastance and to predict arterial pressure response to fluid administration. Anesth Analg 2015;120:76-84.

45. guinot Pg, Bernard e, levrard M, Dupont H, lorne e. Dynamic arterial elastance predicts mean arterial pres-sure decrease associated with decreasing norepinephrine dosage in septic shock. crit care 2015;19:14.

46. Monge garcía Mi, gil cano a, gracia romero M. Dy-namic arterial elastance to predict arterial pressure re-sponse to volume loading in preload-dependent patients. crit care 2011;15:r15.

47. Vos JJ, Kalmar aF, struys MM, Wietasch JK, Hendriks Hg, scheeren tW. comparison of arterial pressure and plethysmographic waveform-based dynamic preload variables in assessing fluid responsiveness and dynamic arterial tone in patients undergoing major hepatic resec-tion. Br J anaesth. 2013;110:940-6.

48. Wyler von Ballmoos M, takala J, roeck M, Porta F, tueller D, ganter cc, et al. Pulse-pressure variation and hemodynamic response in patients with elevated pulmonary artery pressure: a clinical study. crit care 2010;14:r111.

49. Michard F, richards g, Biais M, lopes M, auler Jo. Using pulse pressure variation or stroke volume vari-ation to diagnose right ventricular failure? crit care. 2010;14:451.

cannot accurately predict fluid responsiveness in pa-tients with intra-abdominal hypertension. crit care Med 2010;38:1824-9.

27. De Backer D, Heenen s, Piagnerelli M, Koch M, Vin-cent JL. Pulse pressure variations to predict fluid respon-siveness: influence of tidal volume. Intensive Care Med 2005;31:517-23.

28. De Backer D, taccone Fs, Holsten r, ibrahimi F, Vincent JL. Influence of respiratory rate on stroke volume varia-tion in mechanically ventilated patients. anesthesiology 2009;110:1092-7.

29. cannesson M, le Manach Y, Hofer cK, goarin JP, lehot JJ, Vallet B, et al. assessing the diagnostic accuracy of pulse pressure variations for the prediction of fluid re-sponsiveness: a “gray zone” approach. anesthesiology 2011;115:231-41.

30. Biais M, ehrmann s, Mari a, conte B, Mahjoub Y, Desebbe o, et al. clinical relevance of pulse pressure variations for predicting fluid responsiveness in mechani-cally ventilated intensive care unit patients: the grey zone approach. crit care 2014;18:587.

31. Benes J, giglio M, Brienza N, Michard F. the effects of goal-directed fluid therapy based on dynamic parameters on post-surgical outcome: a meta-analysis of randomized controlled trials. crit care 2014;18:584.

32. Fischer Mo, coucoravas J, truong J, Zhu l, gérard Jl, Hanouz Jl, et al. assessment of changes in cardiac index and fluid responsiveness: a comparison of Nexfin and transpulmonary thermodilution. acta anaesthesiol scand 2013;57:704-12.

33. colombo r, Marchi a, Borghi B, Fossali t, rech r, castelli a, et al. Pulse Photoplethysmographic analysis estimates the sympathetic activity Directed to Heart and Vessels. anesthesiology 2015;123:336-45.

34. Biais M, cottenceau V, Petit l, Masson F, cochard JF, sztark F. impact of norepinephrine on the relationship between pleth variability index and pulse pressure vari-ations in icU adult patients. crit care 2011;15:r168.

35. loupec t, Nanadoumgar H, Frasca D, Petitpas F, laksiri l, Baudouin D, et al. Pleth variability index predicts fluid responsiveness in critically ill patients. crit care Med 2011;39:294-99.

36. Monnet X, guérin l, Jozwiak M, Bataille a, Julien F, richard c, et al. Pleth variability index is a weak predic-tor of fluid responsiveness in patients receiving norepine-phrine. Br J anaesth. 2013;110:207-13.

37. Fischer Mo, Pelissier a, Bohadana D, gérard Jl, Hanouz Jl, Fellahi Jl. Prediction of responsiveness to an intra-venous fluid challenge in patients after cardiac surgery with cardiopulmonary bypass: a comparison between arterial pulse pressure variation and digital plethysmo-graphic variability index. J cardiothorac Vasc anesth 2013;27:1087-93.

38. Fischer Mo, Pellissier a, saplacan V, gérard Jl, Hanouz Jl, Fellahi Jl. cephalic versus digital plethysmographic

Conflicts of interest.—Matthieu Biais has receive honoraria as a lecturer from edwards lifescience and Pulsion Medical systems. all other authors declare that they have no conflict of interest regarding the material discussed in this manuscript.Article first published online: July 13, 2016. - Manuscript accepted: July 11, 2016. - Manuscript revised: June 1, 206. - Manuscript received: February 3, 2016.


© 2016 eDiZioni Minerva MeDicaonline version at http://www.minervamedica.itMinerva anestesiologica 2016 october;82(10):1122

Near-zero difficult tracheal intubation and tracheal intubation failure: too good to be true

Dear editor,

We read with interest the article published by Cag-nazzi et al.1 on the successfully implementation of Bes-ta Airway Algorithm in a morbidly obese population, but we would like to point out some concerns about the final message given to the readers of this study:

1. Authors state that there was mandatory need of expert Glidescope (GVL) users for obese patients, and not simply “standard” GVL users. They someway fix expertise level as >20 GVL uses, whereas paper from cortellazzi et al. clearly identifies in more than 70 uses the minimum learning curve for GVL.2 This is probably one of reasons why the first attempt success rate report-ed by Cagnazzi is 87%, that is lower than results re-ported by Yedemann et al.3 in a similar population study. This is, in our opinion, a very important point to clarify before implementation of GVL in routine daily practice.

2. Authors use the IDS and a CL >3 as marker of difficult tracheal intubation (DTI), while we think this is a common misunderstanding, first of all because CL grading was designed for direct laryngoscopy and is not automatically adaptable to indirect (“look around the corner”) laryngoscopy, and second because with vide-olaryngoscopes it is not so rare to observe the “I can see but I can’t intubate scenario”, meaning that intubation could be challenging if not impossible despite a good laryngeal view. On this point of view, using a specific videolaryngoscopy score as that proposed by Freman-tle 4 could be more objective, allowing not overestimate the videolaryngoscopy efficacy and thus resulting safer.

3. The incidence of Difficult Mask Ventilation (DMV) was 8% (17 patients), as reported by literature,5 with transient desaturation in 7 patients (3.2%). We think that this is the main limitation of the proposed Besta Airway Algorithm, since El Ganzouri Risk Index does not pre-dict reliably the DMV: any videolaryngoscope (VLS) algorithm should clearly exclude difficult to ventilate patients taking account that no VLS will ever allow oxy-genation/ventilation, which is a mandatory safety issue in predicted difficult airways and an even more impor-tant goal for a safe obese airway management.

In conclusion, this interesting study provides us new data on the endless debate about the role of videolaryngo-

scope in difficult airway management, alerting everyone for the need to standardize perioperative strategies in or-der to guarantee airway safety management particularly for the obese patient. How can we identify and how can we manage the patients with predicted DMv? How can we define expert an operator? Which back-up device should be used in case of GVL failure? These are just some of the questions to be answered by the forthcoming studies.

Ruggero M. CORSO 1*, ida Di giacinto 2, Massimiliano SORBELLO 3,

Emanuele PIRACCINI 1, Flavia PETRINI 4

1Departement of Surgery, Anesthesia and Intensive Care Section, “GB Morgagni-L. Pierantoni” Hospital, Forlì,

Italy; 2anesthesia and intensive care, azienda ospedaliero-Universitaria Policlinico Sant’Orsola-Malpighi, Bologna,

Italy; 3Anesthesia and Intensive Care AOU Policlinico Vittorio Emanuele Catania, Catania, Italy; 4Department of Perioperative Medicine, Pain, ICU and RRS, Chieti

University Hospital, ASL 2 Abruzzo, Chieti Pescara, Italy

*Corresponding author: Ruggero M. Corso, Department of Sur-gery, Anesthesia and Intensive Care Section, “GB Morgagni-L. Pierantoni” Hospital, Viale Forlanini 34, 47100, Forlì, Italy. e-mail: [email protected]

References

1. Cagnazzi E, Mosca A, Pe F, Togazzari T, Manenti O, Mit-tempergher F, et al. Near-zero diffi cult tracheal intuba-Near-zero difficult tracheal intuba-tion and tracheal intubation failure rate with the “Besta Airway Algorithm” and “Glidescope® in morbidly obese” (GLOBE). Minerva Anestesiol 2016;82:966-73.

2. Cortellazzi P, Caldiroli D, Byrne A, Sommariva A, Orena EF, Tramacere I. Defining and developing expertise in tracheal intubation using a GlideScope(®) for anaesthet-ists with expertise in Macintosh direct laryngoscopy: an in-vivo longitudinal study. Anaesthesia 2015;70:290-5.

3. Ydemann M1, Rovsing L, Lindekaer AL, Olsen KS. In-tubation of the morbidly obese patient: GlideScope(®) vs. Fastrach™.Acta Anaesthesiol Scand 2012;56:755-61.

4. Swann AD, English JD, O’Loughlin EJ. The develop-ment and preliminary evaluation of a proposed new scor-ing system for videolaryngoscopy. Anaesth Intensive Care 2012;40:697-701.

5. Leoni A, Arlati S, Ghisi D, Verwej M, Lugani D, Ghi-si P, et al. Difficult Mask Ventilation in obese patients: analysis of predictive factors. Minerva Anestesiol 2014;80:149-57.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: May 14, 2016. - Manuscript ac-cepted: May 12, 2016. - Manuscript received: April 22, 2016.(Cite this article as: Corso RM, Di Giacinto I, Sorbello M, Pi-raccini E, Petrini F. Near-zero difficult tracheal intubation and tracheal intubation failure: too good to be true. Minerva Aneste-siol 2016;82:1122)

L E T T E R S T O T H E E D I T O R

anno: 2016Mese: octoberVolume: 82no: 10rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

Lavoro: 11397-MAStitolo breve: primo autore: LETTERS TO THE EDITORpagine: 1122-000citazione: Minerva anestesiol 2016;82:1122-000

letters to tHe eDitor


DMV less likely. Six out of the seven transient desaturations described in our study occurred when an orotracheal tube was already rightly positioned and were therefore rapidly corrected. Our results suggest that the Besta Airway Algo-rithm can be safely implemented in clinical practice. That said, we agree with corso et al. that every effort to further reduce the risk of DMV should be made and future studies directed at strenghthening the induction strategy (e.g. a phar-macological approach for maintain spontaneous breathing) should be undertaken.

elena cagnaZZi 1 *, nicola latronico 2, Federico PE 1, alessandro Mosca 1

1Department of anaesthesia, critical care Medicine and Emergency, University of Brescia

at Spedali Civili, Brescia, Italy;2 University of Brescia at Spedali Civili, Brescia, Italy

Corresponding author: Elena Cagnazzi, Department of Anes-thesia, Critical Care Medicine and Emergency, Azienda Ospe-daliera Spedali Civili di Brescia, Piazzale Ospedali Civili 1, 25123 Brescia, Italy. e-mail: [email protected]

References

1. Corso RM, Di Giacinto I, Sorbello M, Piraccini E, Petrini F. Near-zero difficult tracheal intubation and tracheal in-tubation failure: too good to be true. Minerva Anestesiol 2016;82:1122.

2. Cagnazzi E, Mosca A, Pe F, Togazzari T, Manenti O, Mit-tempergher F, et al. Near-zero difficult tracheal intubation and tracheal intubation failure rate with the “Besta Air-way Algorithm” and “Glidescope® in morbidly obese” (GLOBE). Minerva Anestesiol 2016;82:966-73

3. Cortellazzi P, Caldiroli D, Byrne A, Sommariva A, Orena EF, Tramacere I. Defining and developing expertise in tracheal intubation using a GlideScope(®) for anaesthe-tists with expertise in Macintosh direct laryngoscopy: an in-vivo longitudinal study. Anaesthesia 2015;70:290-5.

4. Correa JB, Broenstrup, Dellazzana JE, Flores, Sturm A, Moore D, et al. Aplicação da Curva CUSUM para Avaliar o Treinamento da Intubação Orotraqueal com o Laringoscópio Truview EVO2 ® * Using the Cusum Curve to Evaluate the Training of Orotracheal Intubation with the Truview EVO2 ® Laryngoscope. Rev Bras Anestesiol 2009;59:321-31.

5. Tachibana N, Niiyama Y, Yamakage M. Incidence of can-not intubate-cannot ventilate (CICV): results of a 3-year retrospective multicenter clinical study in a network of university hospitals. J Anesth 2015;29:326-30.

6. Nicholson A, Smith A, Lewis S, Cook T. Tracheal intuba-tion with a flexible intubation scope versus other intubation techniques for obese patients requiring general anaesthesia (review). cochrane libr cochrane collab 2014;1:1-45.

7. Kheterpal S, Martin L, Shanks AM, Tremper KK. Predic-tion and outcomes of impossible mask ventilation: a review of 50,000 anesthetics. Anesthesiology. 2009;110:891-7.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: July 26, 2016. - Manuscript ac-cepted: July 25, 2016. - Manuscript received: June 9, 2016.(Cite this article as: Cagnazzi E, Latronico N, Pe F, Mosca A. Besta Airway Algorithm in morbidly obese patients: expertise and safety. Minerva Anestesiol 2016;82:1123)

© 2016 eDiZioni Minerva MeDicaonline version at http://www.minervamedica.it ; ( ):

Besta Airway Algorithm in morbidly obese patients: expertise and safety

Dear editor,

We thank Dr. Corso et al.1 for their comments, which give us the opportunity to clarify some of the results of our study.2

We agree about the importance of adequate training before implementation of Glidescope videolaryngoscope (GVL) in clinical practice. When our study began, the Cor-tellazzi’s paper 3 had not been published yet: we used the CUSUM curve methodology which attests the expert status of the operator 4 objectively. The Yedemann study was not powered to detect differences in first time intubation suc-cess. In fact, differences in first time intubation success be-tween GVL and Fastrach were not statistically significant (92% versus 84%, P=0.22). Therefore, it is highly unlikely that the differences between Yedemann results and ours (87%) be significant.

We agree that the use of C&L alone can be misleading. Conversely, the concurrent documentation of C&L, IDS, the device used, and the alternative solutions adopted, pro-vides the same information, if not more, as the Fremantle score, which, to the best of our knowledge, has been evalu-ated only preliminarily in non-obese simulation settings. The “I can see but I can’t intubate” scenario typically takes place with a “non-expert/non-routine GVL user”. The Bes-ta Airway Algorithm rescue plan consists in not reiterating the intubation attempts,5 oxygenating and awakening the patient: the habit of declaring failure of intubation early on, irrespective of the glottis view, and of moving early to laryngeal mask ventilation, are the pillars of a safe patient awakening. The fact that the videolaryngoscope (VLS) looks ‘around the corner’ implies a better glottis view and a different technique for intubation but the anatomical struc-tures to describe remain the same. The C&L score simply aids with the classification of the visible structures.

We agree that a strategy to prevent difficult mask ventila-tion (DMV) has to be put in place but which is the safest way has yet to be established.6 the available predictors for DMv and iMv 7 do not reach sufficient accuracy, and hence it is difficult to establish a threshold value without generat-ing a high number of false positives: to schedule these pa-tients for awake-fiberoptic intubation might not be always a feasible nor an adequate approach. In the Besta Airway Algorithm several modalities of DMV/desaturation pre-vention are implemented, among which the exclusion of patients with neck abnormalities, the awake-fiberoptic intu-bation, the measure of not reiterating intubation attempts 5 and the prompt institution of laryngeal mask ventilation. It goes without saying that VLS does not oxygenate, however, the high probability of an uneventful GVL intubation makes

Minerva anestesiologica 2016 october;82(10):1123



ver) was maintained for 120 minutes, reaching 35 °C intraoperative esophageal temperature. Neuromuscular blockade was reversed with atropine and neostigmine, and the patient was transferred awake to the recovery room in a 30° head-up position, with ongoing 50 mL/h fluids after receiving 1500 mL intraoperatively.

After 20 minutes the patient became acutely agi-tated, tachypneic and hypotensive (60/45 mmHg), developing venous congestion, atrial fibrillation (150 beats/min), ECG repolarization abnormalities (Fig-ure 1), and metabolic acidosis (pH 7.22; PaO2 100 mmHg; PaCO2 32 mmHg; Hco3

– 18 mEq/L; BE -13 mmol/L). Echocardiography showed right ventricular distension, diffuse hypokinesia, and atrial septal left-ward bulging but no shunts. Central venous pressure was 15 mmHg. Resuscitation comprised 100% oxy-gen and intravenous epinephrine infusion (0.1 µg/kg/min). Conditions progressively resolved and epineph-rine was discontinued after 3 hours, with normalized ECG, ventricular function, and arterial and central venous pressures. Angiography ruled out coronary disease. Troponin I was mildly elevated (peak 0.26 ng/L) and the patient was discharged 15 days postop-eratively with no sequelae.

High solubility and rapid absorption renders CO2 the most employed gas to induce pneumoperitoneum for laparoscopy. Volume and timing of CO2 absorption and elimination capacity determine embolism severity, ultimately manifesting as a “gas lock“ in the pulmonary circulation, with acute right ventricular afterload mis-match and circulatory collapse. Positive peritoneal pres-sure may enhance CO2 entering the circulation through


Delayed CO2 embolism: importance of early postoperative surveillance

Dear editor,

carbon dioxide (co2) peritoneal insufflation is daily-performed for laparoscopy. Intravascular CO2 entrapment, and subsequent embolization, is clinically apparent in 0.0014-0.6% of the cases, but fatal in 30%.1 Conversely, 5-30% of perioperative complications, of-ten requiring intensive care, occur in ASA physical sta-tus 1-2 patients.2 We present a case of delayed CO2 em-bolism following laparoscopy, manifesting outside the operative theatre. Consent for scientific purposes was obtained with informed consent for anesthesia.

A 61-year-old woman (height 150 cm, weight 45 kg; ASA Score 2) with malignancy and no cardiovascular risk factor underwent laparoscopic right hemicolecto-my. No instability ensued during anesthesia induction (2 mg/kg propofol, 2 µg/kg fentanyl) or maintenance (6% desflurane). Atracurium (induction, 0.5 mg/kg; 10 mg boluses, as needed) provided muscle relaxation. co2 insufflation was initiated in the supine patient, at-taining a 15-mmHg intraabdominal pressure. A Tren-delenburg and left lateral decubitus (Durant’s manoeu-

Figure 1.—ECG recorded immediately after circulatory collapse, showing atrial fibrillation, diffuse S-T segment depression, markedly in the anterolateral leads, primarily suggesting diffuse sub-endocardial ischemia in the left coronary territory.

Minerva anestesiologica 2016 october;82(10):1124-5




Cervical emphysema and pneumomediastinum due to isolated pharyngeal perforation after blunt trauma

Dear editor,

Pharyngeal lesion after blunt trauma is a rare oc-currence.1 Nevertheless, early diagnosis is essential in order to prevent life-threating complications, like retro-pharyngeal abscess, mediastinitis and airway compro-mise. In this paper we report our management of a case of hypopharyngeal perforation after blunt trauma.

After a car accident, a 21-year-old man was trans-ferred to the Emergency Department. He reported hitting his head, neck and chest on the steering wheel and wind-screen of the car because he did not wear seat belt and the air bags did not open. In the first examination, his vital signs were stable, but he was found to have neck crepitus and shallow abrasions. However, no evidence of pene-trating trauma emerged. He suffered from dyspnea, neck pain, odynophagia and hemoptysis. A computed tomog-raphy (CT) was performed and it demonstrated cervical emphysema, massive pneumomediastinum and a lesion of the posterior wall of the hypopharynx extending for 2.5 cm, in correspondence of pharyngo-esophageal junc-tion (Figure 1). No foreign bodies were identified nor were other lesions found in the body. A bronchoscopy revealed a normal larynx and tracheobrochial tree. Be-cause of ingravescent cervical emphysema and dyspnea, the patient required endotracheal intubation and the admission to our Intensive Care Unit (ICU). Antibiotic therapy was started and a nasogastric tube was inserted, under laryngoscopy vision, for enteral feeding. Because of the size of the lesion and the involvement of the upper esophagus, we opted for transorally surgical repair. After two days, a neck and chest radiogram revealed reabsorp-tion of air and the patient was successfully extubated and discharged from our ICU without sequelae.

Injury to the pharynx most commonly occurs after instrumentation, such as difficult intubation or head and neck surgery, or foreign body ingestion. Trau-matic pharyngeal lesion can be the consequence of a penetrating or blunt trauma: the latter is a very rare cause of pharyngoesophageal injury (<1%).2 In our case, as no signs of penetrating trauma were identified, we can speculate that a sudden pressure increase in the pharynx during blunt trauma of the neck might have caused tearing of the posterior hypopharyngeal wall (barotrauma). Alternatively, the hyperextension and flexion of the neck, after fast deceleration, may have pushed the pharyngo-esophageal junction against the

an injured vessel, whereas portal system gas entrapment is supposed to account for later onset embolization.3 Paradoxical embolism from a systemic vein has also been described, typically through a patent foramen ova-le, but also across physiological intrapulmonary shunts with no evidence of intracardiac shunts.4

We speculated that CO2 initially entered the blood-stream during removal of the resected colon through the small incision with stretched tissues and a CO2-pressurized peritoneal cavity. The steep head-down and left lateral decubitus displaced gas predominantly to the right and caudally, determining temporary intravascular entrapment in the portal circulation or, possibly, right ventricular apex, and subsequent pulmonary migration only after the patient was positioned in a 30° head-up decubitus, explaining later onset presentation.5

In conclusion, life-threatening CO2 embolism may present with delay during early hours following lapa-roscopy, which emphasizes the critical importance of postoperative surveillance despite minor comorbidities.

albino leoni 1 *, andrea antonelli 2, Marco POCAR 2, 3

1Deparment of Anesthesiology and Intensive Care,IRCCS MultiMedica, Sesto San Giovanni, Milan, Italy;

2Cardiovascular Department, IRCCS MultiMedica,Sesto San Giovanni, Milan, Italy;

3Università degli Studi di Milano, Milan, Italy*Corresponding author: Albino Leoni, Department of Anesthe-siology and Intensive Care, IRCCS MultiMedica, via Milanese 300, I-20099 Sesto San Giovanni, Milan, Italy. e-mail: [email protected]

References

1. Bonjer HJ, Hazebroek EJ, Kazemier G, Giuffrida MC, Meijer WS, Lange JF. Open versus closed establishment of pneumoperitoneum in laparoscopic surgery. Br J Surg 1997;84:599-602.

2. Kluger MT, Bullock MFM. Recovery room incidents: a review of 419 reports from the anaesthetic incident moni-toring study (AIMS). Anaesthesia 2002;57:1060-6.

3. Gutt CN, Oniu T, Mehrabi A, Schemmer P, Kashfi A, Kraus T, et al. Circulatory and respiratory complications of carbon dioxide insufflation. Dig Surg 2004;21:95-105.

4. Thackray NM, Murphy PM, McLean RF, deLacy JL. Ve-nous air embolism accompanied by echocardiographic evidence of transpulmonary air passage. Crit Care Med 1996;24:359-61.

5. Mirski MA, Lele AV, Fitzsimmons L, Toung TJ. Diagno-sis and treatment of vascular air embolism. Anesthesiol-ogy 2007;106:164-77.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: May 14, 2016. - Manuscript ac-cepted: May 12, 2016. - Manuscript revised: April 19, 2016. - Manuscript received: December 17, 2015.(Cite this article as: Leoni A, Antonelli A, Pocar M. Delayed co2 embolism: importance of early postoperative surveillance. Minerva anestesiol 2016;82:1124-5)

Minerva anestesiologica 2016 october;82(10):1125-6



surgical repair should be considered for large perfora-tion and for tears that extend into the esophagus.4

Matteo vissani 1 *, liana lentiscHio 1, giammichele nicoletta 1,

Fabrizio PUGLIESE 2

Costanzo FEDELI 3, raffaele Zava 1

1Department of anesthesia, intensive Care and Pain Medicine,

San Giovanni Battista Hospital, Foligno, Italy;2Department of Radiology, San Giovanni

Battista Hospital, Foligno, Italy;3Department of Thoracic Surgery,

San Giovanni Battista Hospital, Foligno, Italy*Corresponding author: Matteo Vissani, Department of Anes-thesia, Intensive Care and Pain Medicine, “San Giovanni Bat-tista” Hospital, Via Massimo Arcamone, 06034 Foligno (PG), Italy. E-mail: [email protected]

References

1. Hagr A, Kamal D, Tabah R. Pharyngeal perforation caused by blunt trauma to neck. Can J Surg 2003;46:57-8.

2. Demetriades D, velmahos gg, asensio Ja. cervical pharyngoesophageal and laryngotracheal injuries. World J Surg. 2001;25:1044-8.

3. Stringer WL, Kelly DL, Johnston Fr, Holliday RH. Hy-perextension injury of the cervical spine with esophageal perforation. J Neurosurg 1980;53:541-3.

4. Niezgoda JA, McMenamin P, Graeber GM. Pharyn-goesophageal perforation after blunt neck trauma. Ann Thorac Surg 1990;50:615-7.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: May 27, 2016. - Manuscript ac-cepted: May 25, 2016. - Manuscript revised: May 13, 2016. - Manuscript received: March 15, 2016.(Cite this article as: Vissani M, Lentischio L, Nicoletta G, Pug-liese F, Fedeli C, Zava R. Cervical emphysema and pneumo-mediastinum due to isolated pharyngeal perforation after blunt trauma. Minerva Anestesiol 2016;82:1125-6)

cervical spine, thus entrapping the hypopharyngeal wall between the vertebral bodies as hyperextension changed into flexion.3 In this case, the immediate intu-bation of the trachea isolated the pharyngeal lesion and prevented further impairment of the airway and prompt antibiotic therapy together with surgical repair avoided mediastinitis. The prognosis of pharyngeal and cervical esophageal injuries depends on many factors, such as associated injuries, shock on admission, mechanism of injury and timing of operation:2 in our case prognosis was good because of the early detection and treatment of the lesion and the mechanism of injury (blunt trau-ma). as regards the treatment, conservative manage-ment with antibiotics and fasting can be reserved for small perforation (<2 cm) limited to the pharynx while

Figure 1.—CT scan of the neck: white arrow indicates the pharyngeal lesion. Note the surrounding cervical emphysema.

Vol. 82 - No. 10 MINERVA ANESTESIOLOGICA 1127

T O P 5 0 M I N E R V A A N E S T E S I O L O G I C A R E V I E W E R S

Anno: 2016Mese: OctoberVolume: 82No: 10Rivista: MINERVA ANESTESIOLOGICACod Rivista: Minerva Anestesiol

Lavoro: titolo breve: primo autore: ACKNOWLEDGEMENTSpagine: 1127

Top 50 reviewers March 2016-August 2016

Surname Name Numberof revisions

Savoia Gennaro 11Biasucci Daniele Guerino 11Camporesi Enrico Mario 10Agrò Eugenio Felice 9Ezzeldin Ibrahim Saleh 9Cattano Davide 9Benes Jan 8Bertini Pietro 7Aceto Paola 7Donadello Katia 7Alston Theodore 6El Tahan Mohamed 6Abad Gurumeta Alfredo 6Faraoni David 6Caruselli Marco 6Onutu Adela Hilda 5Brogly Nicolas 5Dalfino Lidia 5Gómez-Ríos Manuel Ángel 5Deflandre Eric P. 5Brazzi Luca 5Gottumukkala Vijaya Narasimha Raju 4Tritapepe Luigi 4Richa Freda 4Montini Luca 4Max Martin 4Vaida Sonia 4Ting Chien-Kun 4Turnbull David 4Arnal Jean-Michel 4Kayhan Zeynep 4Sbaraglia Fabio 4Guarracino Fabio 4Kinoshita Hiroyuki 3Bruder Nicolas 3De Blasi Roberto Alberto 3Peris Adriano 3Diaz-Gomez Jose L. 3Riem Nicole 3Cavallone Laura Francesca 3Kim Tae W. 3De Pascale Gennaro 3Aly Essam 3Clendenen Steven R 3Kakodkar Prashant Shivaji 3Donatelli Francesco 3Beilin Yaakov 3Cinnella Gilda 3Borghi Battista 3Somaini Marta 3

VOLUME 82 · No. 10 · OCTOBER 2016...ECMO management guided by echo: is it feasible? Tritapepe L....

Documents

Transcript of VOLUME 82 · No. 10 · OCTOBER 2016...ECMO management guided by echo: is it feasible? Tritapepe L....